This paper discusses uncertainties in software develop ment for reuse and maintenance. In particular, we present how fuzzy techniques can help in handling a source of uncertainty: the classification of components and their retrieval for reuse according to software behavioral properties. Behavioral classification is inherently impre.cise, due to the fact that any components may exhibit several behaviors, depending on the application viewpoint. A model is described based on a repository where software descriptors are stored. Descriptors contain lexical elements, characterizing software behavior, weighted using fuzzy sets. Fuzzy weighting expresse the imprecision of behavioral descriptions, and allows one to explore the repository via imprecise queries. The retrieval has adaptive capabilities based on observation by the system of users' choices of candidate components.
Fuzzy techniques for software reuse / E. Damiani, M.G. Fugini - In: Applied computing 1996 : proceedings of the 1996 ACM symposium on applied computing : Philadelphia, Pennsylvania, february 17-19, 1996 / [a cura di] K.M. George ... [et al.]. - New York : Association for computing machinery, 1996. - ISBN 0897918207. - pp. 552-557 (( Intervento presentato al 11. convegno ACM Symposium on Applied Computing (SAC) tenutosi a Philadelphia nel 1996 [10.1145/331119.331453].
Fuzzy techniques for software reuse
E. DamianiPrimo
;
1996
Abstract
This paper discusses uncertainties in software develop ment for reuse and maintenance. In particular, we present how fuzzy techniques can help in handling a source of uncertainty: the classification of components and their retrieval for reuse according to software behavioral properties. Behavioral classification is inherently impre.cise, due to the fact that any components may exhibit several behaviors, depending on the application viewpoint. A model is described based on a repository where software descriptors are stored. Descriptors contain lexical elements, characterizing software behavior, weighted using fuzzy sets. Fuzzy weighting expresse the imprecision of behavioral descriptions, and allows one to explore the repository via imprecise queries. The retrieval has adaptive capabilities based on observation by the system of users' choices of candidate components.
Pubblicazioni consigliate
I documenti in IRIS sono protetti da copyright e tutti i diritti sono riservati, salvo diversa indicazione.
