Subproject 2 - Pre-Conflict Decision Tools for Legal Practitioners

Start - 2018 | End 2024 Duration - 6 years

The objective of this Subproject is to analyze the contribution of algorithmic tools capable of predicting the probable outcome of a trial. The first step towards making them useful tools for practitioners is to understand how they work and thus justify their reliability.

Research activities 

Case studies 

The JusticeBot project (Procezeus). This chatbot project is originally aimed at non-attorneys. The researchers intend to develop a rule-tree for the domain, which more naturally would be aimed at attorneys.

This case study bridges Subprojects 1 and 2.

The Case summarization project. A machine learning algorithm will be trained to generate summaries based on a unique training set constituted of pairs of full case reports and summaries prepared by expert humans.



Kevin Ashley

Radboud Winkels

Leila Kosseim






Ministère de la justice





This content has been updated on 18 February 2020 at 18 h 39 min.