SOFTWARE SYSTEM FOR LEGAL DOCUMENT PROCESSING BASED ON SEMANTIC WEB AND MACHINE LEARNING TECHNIQUES

Abstract

This paper presents a software system for processing legal documents of the Australian Federal Court. The mentioned system consists of three logical segments: semantic web, summarization, and topic modeling. The first segment represents the popularization of the ontology and the extraction of relevant semantic connections within it. The second is based on summarizing extensive documents into a short text that contains the most important from the original. The third unit classifies all documents into a defined number of topics to make it easier for the user to search for relevant ones. The proposed system based on the above techniques provides support for semantic search, a brief description of an extensive document, and a display of all documents on the same topic.