Semantics and Law

Le Groupe de Travail TALSEM, financé par DigiCosme propose de rencontrer les professeurs Ken Satoh et Yoshinobu Kano lors du prochain séminaire le mardi 19 mars 2019 - LIMSI - Batiment 507 - 91400 Orsay

Programme

14:00 - Ken Satoh, Professor, National Institute of Informatics (NII), Japan since 2001. He has been working on logical foundations of AI for 30 years. He became recently interested in application of AI to law domain and studied at law schools. He is now developing a system, PROLEG, which simulate judge's reasoning under incomplete information environment.

Title : The current trend of AI applications to legal domain and PROlog-based LEGal reasoning system: PROLEG

Abstract: We firstly give a survey of the current trend of AI applications to legal domain and then give an overview of our project PROLEG where we implement "Japanese Ultimate Fact (JUF) thery" to simulate judge's reasoning at a civil court. JUF theory is a tool for judges to make a judgement based on burden of proof under incomplete information.PROLEG consists of general rules and exceptions which directly reflect lawyers' knowledge structure in legal reasoning.

14h45 - Yoshinobu Kano, Associate Professor, Faculty of Informatics, Shizuoka University, Japan - BS in physics (2001), MSc (2003) and PhD (2011) in information science and technology from the University of Tokyo, respectively. A research associate in University of Tokyo (2009), JST PRESTO researcher (2011), an associate professor (PI) in Shizuoka Univerity (2014-). He is interested in natural language processing and its applications.

Title : Legal, Medical, and Conversational Natural Language Processing: Challenges in Question Answering, Dialog System and Automatic Judgement/Diagnosis Support

Abstract: I would like to introduce our different ongoing projects related to NLP, which I involve as organizers/PIs. Collaborating with Prof. Satoh, we have been holding our annual COLIEE competition style workshop where participants try solving Japanese civil law legal bar exam both in English and Japanese languages. In the medical fields, we have been building diagnosis systems for mental diseases and developmental disorders from conversation records, and text mining system for EHRs (Electronic Health Records). Other research issues include a dialog system that plays a conversation game called "Mafia" (or "are you a werewolf"), sentence generations for advertisement copies and news headlines. The last research issue is text mining for neuroscience papers. I wish to discuss possibility to expand some of these issues into other languages than Japanese.

15:30 - Coffee break

16:00 - Meetings and discussions

17:00 - End of meeting