Artificial Intelligence and Law

, Volume 15, Issue 1, pp 1-17

First online:

Automatic Classification of Provisions in Legislative Texts

  • E. FrancesconiAffiliated withITTIG - CNR, Istituto di Teoria e Tecniche dell’Informazione Giuridica - Consiglio Nazionale delle Ricerche Email author 
  • , A. PasseriniAffiliated withDSI - Dipartimento di Sistemi e Informatica, Università di Firenze

Rent the article at a discount

Rent now

* Final gross prices may vary according to local VAT.

Get Access


Legislation usually lacks a systematic organization which makes the management and the access to norms a hard problem to face. A more analytic semantic unit of reference (provision) for legislative texts was identified. A model of provisions (provisions types and their arguments) allows to describe the semantics of rules in legislative texts. It can be used to develop advanced semantic-based applications and services on legislation. In this paper an automatic bottom-up strategy to qualify existing legislative texts in terms of provision types is described.


model of provisions Naïve Bayes classifier SVM classifier text categorization