A model transformation framework to increase OCL usability

Abstract

The usability of a modeling language has a direct relationship with several factors of models constructed with the modeling language, such as time required and accuracy. Object Constraint Language (OCL) is the most prevalent language to document system constraints that are annotated in the Unified Modeling Language (UML). OCL is reputed as a modeling language with difficult syntax, and prior knowledge of OCL is needed to use the language. These obstacles result in the low usability of OCL. Therefore, the current research proposes a model to automatically transform system constraints formed in English sentences to OCL specifications. The proposed model is based on the Model-Driven Architecture (MDA) approach. The Linear Temporal Logic (LTL) properties of the proposed model are verified by the Maude model checker. To validate the proposed model and compare it with the existing work, the En2OCL (English2OCL) application is developed. This application is tested by three evaluation metrics: precision, recall, and f-measure. The En2OCL application is further compared with the NL2OCLviaSBVR application, which is the existing work on OCL generation from English sentences. The comparison shows a considerable improvement in precision, recall, and f-measure.