Modelling Contracts Using RuleML

This paper presents an approach for the specification and implementation of e-contracts for Web monitoring. This is done in the setting of RuleML. We argue that monitoring contract execution requires also a logical account of deontic concepts and of violations. Accordingly, RuleML is extended to cover these aspects. 1 Background and Motivation Business contracts are mutual agreements between two or more parties engaging in various types of economic exchanges and transactions. They are used to specify the obligations, permissions and prohibitions that the signatories should hold responsible for and to state the actions or penalties that may be taken when any of the stated agreements is not being met. Given the increasing efforts by organisations to carry out their business via the Internet, it is crucial to model contracts in terms of workflows, so that all relevant tasks of contracts can be described as business processes, where business processes are defined by business rules, statements or policies listed in business contracts or other legal documents that are used by organisations to run the activities, to provide an understanding of how a business operates, and to direct the behaviour of the organisation. Business rules are typically expressed implicitly in contracts and they are also hidden in many programs across clients, applications and database tiers of modern business information systems, which makes it even harder to monitor the expected behaviour of the policies. Business rules are usually applicable at two levels: the business domain and the operational level of an information system. In the business domain, business rules are usually represented implicitly in the form of natural language expressions. This implicit form cannot be applied to information systems; rules must be made operational by transforming them into a declarative (rule) language. This paper focuses on transforming business contract rules from natural language into a machine readable and executable form. In particular, our aim is to implement the contract in a way that allows explicit monitoring of rules by the computer for any case of violation. Based on a given contract scenario, the contract, in its human-oriented form, will be analysed and represented in a logical form using Deontic and Defeasible Logic. From its logical form, the contract will be transformed into a machine readable rule notation, based on RuleML, and implemented as executable semantics.