The following study introduces “Hybrid Artificial Intelligence Methodology for Legal Analysis” (HAIMLA). It consists of a six-step method to design, develop, validate and deploy artificial intelligence (AI) systems for legal analyses that are built on asynchronous unsupervised and supervised techniques applied to legal texts serialised in the Akoma Ntoso XML standard. HAIMLA methodology is drawn upon the existing literature and case studies in AI & Law and it is grounded on consolidated philosophical approaches. Taken together, this background inspires design requirements that constitute the essential pillars of HAIMLA and new directions for future refinements and implementations.
Palmirani, M., Sapienza, S., Ashley, K. (2024). A Hybrid Artificial Intelligence Methodology for Legal Analysis. BIOLAW JOURNAL, 3, 389-409 [10.15168/2284-4503-3206].
A Hybrid Artificial Intelligence Methodology for Legal Analysis
Monica Palmirani;Salvatore Sapienza
;Kevin Ashley
2024
Abstract
The following study introduces “Hybrid Artificial Intelligence Methodology for Legal Analysis” (HAIMLA). It consists of a six-step method to design, develop, validate and deploy artificial intelligence (AI) systems for legal analyses that are built on asynchronous unsupervised and supervised techniques applied to legal texts serialised in the Akoma Ntoso XML standard. HAIMLA methodology is drawn upon the existing literature and case studies in AI & Law and it is grounded on consolidated philosophical approaches. Taken together, this background inspires design requirements that constitute the essential pillars of HAIMLA and new directions for future refinements and implementations.File | Dimensione | Formato | |
---|---|---|---|
A Hybrid Artificial Intelligence Methodology for Legal Analysis.pdf
accesso aperto
Tipo:
Versione (PDF) editoriale
Licenza:
Licenza per Accesso Aperto. Creative Commons Attribuzione - Non commerciale - Non opere derivate (CCBYNCND)
Dimensione
710.26 kB
Formato
Adobe PDF
|
710.26 kB | Adobe PDF | Visualizza/Apri |
I documenti in IRIS sono protetti da copyright e tutti i diritti sono riservati, salvo diversa indicazione.