Semantic Mark-up of Italian Legal Texts Through NLP-based Techniques

Proceedings of the Fourth International Conference on Language Resources and Evaluation (LREC 2004)

Abstract

In this paper we illustrate an approach to information extraction from legal texts using SALEM. SALEM is an NLP architecture for semantic annotation and indexing of Italian legislative texts, developed by ILC in close collaboration with ITTIG-CNR, Florence. Results of SALEM performance on a test sample of about 500 Italian law paragraphs are provided.