Discovering business rules through process mining

Abstract

Business rules guide the operation of an organization, thus its documentation provides an important source of information both for developing technological solutions (information systems, databases)and for evaluating information systems implementations. Despite its importance, manual creation and maintenance of business rule documentation is very costly, and practically infeasible in complex organizations. This paper describes a method for discovering business rules from the information systems event logs, through the use of process mining and data mining techniques. We exemplify the method execution to discover two selected sub-types of business rules, namely condition action assertions and authorization action assertions. © 2009 Springer Berlin Heidelberg.