Background. Functional size measurement methods aim at measuring the size of functional user requirements of software applications. Functional user requirements can be represented via different notations, including UML diagrams. Objectives. In this paper, the relationship between functional size measures (namely IFPUG Function Points and COSMIC Function Points) and object-oriented measures of UML diagrams representing functional requirements are investigated. Method. A set of functional requirement specifications was modeled via UML diagrams. The functional size measures of user requirements were derived via the standard IFPUG and COSMIC processes; the corresponding UML models were measured using a set of object-oriented metrics that are applicable to UML models representing requirements. Functional size measures were then compared to object-oriented measures. Results. Statistically significant linear regression models were found. It was also found that object-oriented measures of UML requirements models can be used to estimate functional size measures with good accuracy. Conclusions. The obtained results suggest that object-oriented measures –which tools can automatically extract from UML models– provide indications concerning requirements size that are substantially equivalent to those provided by functional size measures.
A study of the correlation between functional size measures and object-oriented measures from UML requirements models
Lavazza, Luigi;
2018-01-01
Abstract
Background. Functional size measurement methods aim at measuring the size of functional user requirements of software applications. Functional user requirements can be represented via different notations, including UML diagrams. Objectives. In this paper, the relationship between functional size measures (namely IFPUG Function Points and COSMIC Function Points) and object-oriented measures of UML diagrams representing functional requirements are investigated. Method. A set of functional requirement specifications was modeled via UML diagrams. The functional size measures of user requirements were derived via the standard IFPUG and COSMIC processes; the corresponding UML models were measured using a set of object-oriented metrics that are applicable to UML models representing requirements. Functional size measures were then compared to object-oriented measures. Results. Statistically significant linear regression models were found. It was also found that object-oriented measures of UML requirements models can be used to estimate functional size measures with good accuracy. Conclusions. The obtained results suggest that object-oriented measures –which tools can automatically extract from UML models– provide indications concerning requirements size that are substantially equivalent to those provided by functional size measures.I documenti in IRIS sono protetti da copyright e tutti i diritti sono riservati, salvo diversa indicazione.