Publication | Closed Access
Capturing more world knowledge in the requirements specification
120
Citations
13
References
1982
Year
Unknown Venue
EngineeringRequirement ModelingSoftware EngineeringMore World KnowledgeSemantic WebSemanticsSoftware AnalysisSoftware RequirementNon-functional RequirementRequirements EngineeringKnowledge RepresentationSoftware RequirementsRequirement EngineeringDesignRequirements ModelsSoftware DesignAutomated ReasoningProgram AnalysisSoftware TestingFormal MethodsRequirements ModelingFunctional SpecificationsKnowledge ManagementFunctional RequirementData Modeling
Software requirements are viewed as representations of substantial real‑world knowledge, not merely functional specifications. The paper presents a Requirements Modeling Framework (RMF) and illustrates its main features. RMF records information about objects, activities, and assertions uniformly via properties, groups entities into classes and metaclasses, and organizes them into generalization hierarchies to support classification, aggregation, and generalization abstraction principles. By providing a mathematical model underlying the terminology, RMF achieves unambiguity and enables consistency verification of the model.
The view is adopted that software requirements involve the representation (modeling) of considerable real-world knowledge, not just functional specifications. A framework (RMF) for requirements models is presented and its main features are illustrated. RMF allows information about three types of conceptual entities (objects, activities, and assertions) to be recorded uniformly using the notion of properties. By grouping all entities into classes or metaclasses, and by organizing classes into generalization (specialization) hierarchies, RMF supports three abstraction principles (classification, aggregation, and generalization) which appear to be of universal importance in the development and organization of complex descriptions. Finally, by providing a mathematical model underlying our terminology, we achieve both unambiguity and the potential to verify consistency of the model.
| Year | Citations | |
|---|---|---|
Page 1
Page 1