Title: Detecting and Explaining Unfairness in Consumer Contracts with Memory Networks

Abstract: We present an approach based on Memory-Augmented Neural Networks for the task of detecting clauses in online Terms of Service that are potentially unfair for the consumer, with the additional goal to explain the legal rationale behind unfairness. The proposed approach goes in the direction of developing neural models for the legal domain, whose output is interpretable and explainable by humans.

Main author bio: Federico Ruggeri was born in Bologna, Italy, 1993. He received Bachelor of Science and Master degrees in Computer Science and Engineering at the University of Bologna and successively joined as a PhD student the department of Computer Science and Engineering (DISI) under the supervision of Prof. Paolo Torroni (Unibo) and PhD Marco Lippi (Unimore). His main research focus is in artificial intelligence (AI) and natural language processing (NLP). In particular, he's currently specializing in Argumentation Mining, a sub-branch research area of NLP dedicated to the automatic extraction of arguments and inferring potential argument-to-argument links. In order to develop innovative solutions, he takes inspiration from other research areas, such as Reinforcement Learning and Geometric Deep Learning.