Change-Aware Legal Document Retrieval Model

No Thumbnail Available



Journal Title

Journal ISSN

Volume Title


Addis Ababa University


Legal documents play a basic role in discharging the law to the public, besides constituting learning material for students, researchers and legal practitioners. Several web portals and document repositories are hosting legal documents. Legal documents contain text rich contents that can be structured and marked with description languages such as XML. This basic feature can be exploited by XML based retrieval models to return legal document parts as a match for information needs rather than presenting the entire statue. The main principle behind the XML based retrieval models is to take the structured information of documents (as formatted by XML tags) and the hierarchical document organization into account when responding to user queries and when computing the relevance ranking. Under different paradigms, legal documents are subjected to change, as the laws they carry undergo modifications or amendments. The dynamicity of the law is manifested by the change in the legal documents, and this change in turn will have a sever consequence on the retrieval quality of XML based legal document retrieval systems. In this work we present a generic model for a change-aware legal document retrieval system using XML based IR approach. Based on the model designed we have developed a prototype system that demonstrates the validity of our ideas and algorithms. By identifying the status of legal contents and by tracking their relationship, we have achieved an improved access to legislative contents. In order to realize this approach we have incorporated the Lucene searching and indexing APIs and proposed a novel re-ranking algorithm that considers the changes in the legal document contents and status of the law contained. The evaluation of our system shows a significant turnover as the result of incorporating the idea of re-ranking search results. Before re-ranking the average precision value for the first five results returned was 46% but after re-ranking the average precision becomes 82.5% in satisfying user information needs.



Change-Aware Legal Document Retrieval Model, Xml Based Retrieval Model, Legal Document Retrieval, Re-Ranking Search Results in Legal Document Retrieval Systems