In this paper, we propose a bootstrapping approach for semi- A utomated legal knowledge extraction. The approach is characterized by the use of a reference legal ontology that is progressively enriched with relevant concepts and related terms extracted from a corpus of le-gal documents (i.e., Court Decision documents). Supervised, multi-label classification techniques and black-box model explanation techniques are the core components of the bootstrapping approach i) to associate CD documents with appropriate concepts in the ontology and ii) to choose the terms that are decisive for determining the association between a document and a certain ontology concept, respectively. The goal of the proposed approach is to reduce the manual involvement of legal experts as much as possible and to improve the accuracy of document classifica-tion, by progressively enriching the term sets associated with ontology concepts. Preliminary experimental results are finally provided to show the contribution of the proposed approach on a corpus of real Court Decision documents.
A Bootstrapping Approach for Semi-Automated Legal Knowledge Extraction and Enrichment / S. Castano, M. Falduti, A. Ferrara, S. Montanelli (CEUR WORKSHOP PROCEEDINGS). - In: SEBD 2020 : Italian Symposium on Advanced Database Systems / [a cura di] M. Agosti, M. Atzori, P. Ciaccia, L. Tanca. - [s.l] : CEUR-WS, 2020. - pp. 1-11 (( Intervento presentato al 28. convegno Italian Symposium on Advanced Database Systems tenutosi a Villasimius nel 2020.
|Titolo:||A Bootstrapping Approach for Semi-Automated Legal Knowledge Extraction and Enrichment|
|Parole Chiave:||legal ontology; legal knowledge extraction; automated Court-Decision analysis|
|Settore Scientifico Disciplinare:||Settore INF/01 - Informatica|
|Data di pubblicazione:||2020|
|Tipologia:||Book Part (author)|
|Appare nelle tipologie:||03 - Contributo in volume|