Compliance checking on first-order knowledge with conflicting and compensatory norms: a comparison among currently available technologies

Abstract

This paper analyses and compares some of the automated reasoners that have been used in recent research for compliance checking. Although the list of the considered reasoners is not exhaustive, we believe that our analysis is representative enough to take stock of the current state of the art in the topic. We are interested here in formalizations at the first-order level. Past literature on normative reasoning mostly focuses on the propositional level. However, the propositional level is of little usefulness for concrete LegalTech applications, in which compliance checking must be enforced on (large) sets of individuals. Furthermore, we are interested in technologies that are freely available and that can be further investigated and compared by the scientific community. In other words, this paper does not consider technologies only employed in industry and/or whose source code is non-accessible. This paper formalizes a selected use case in the considered reasoners and compares the implementations, also in terms of simulations with respect to shared synthetic datasets. The comparison will highlight that lot of further research still needs to be done to integrate the benefits featured by the different reasoners into a single standardized first-order framework, suitable for LegalTech applications. All source codes are freely available at https://github.com/liviorobaldo/compliancecheckers, together with instructions to locally reproduce the simulations.

An ASP Implementation of Defeasible Deontic Logic

“…Here, we are going to illustrate and evaluate how to use DDL and its ASP encoding with the help of an example taken from [24,21]. The scenario has been recently used to test several rule-based implementations of legal reasoners [43].…”

“…Recently [9] proposed a comparison among different propositional rule-based engines for legal reasoning where SPINdle [34], an implementation of Defeasible (Deontic) Logic, provided the most comprehensive outcome. Similarly, [43] analysed several non-monotonic reasoners (including ASP-based engines and SPINdle) with encodings for legal reasoning, allowing for first-order logic instances. While DDL performed well in the propositional case, there were some concerns about when first-order instances are allowed.…”

“…In Section 2 we provide a quick overview of DDL; in Section 3 we show how to encode the reasoning mechanism of DDL in ASP adopting a meta-program approach, and then we evaluate the encoding in Section 4 against the benchmark developed in [43]. The evaluation will allow us to compare the performance of the novel ASP implementation of DDL and the other ASP encodings of [43]. Finally, we conclude the paper with a quick review of some related work and concluding remarks in Section 5.…”

Defeasible Deontic Logic in Answer Set Programming

“…Here we are going to illustrate how to use L4 and its ASP encoding with the help of an example taken from [9,8]. The scenario has been recently used to test several rule-based implementations of legal reasoners [19].…”

“…In contrast with the ASP-based Defeasible Deontic Logic implementation of L4, none of the implementations analysed in [19] can deal with Article 5 (Automatic Termination). Moreover, none of the other implementations offers a native treatment of the compensatory obligations and has to depend on a contrary-to-duty representation (having to explicitly use obligation A and not A in the antecedent of a new rule).…”