Research And Application Of Correlation Measurement And Abstraction Of Component Based On UCM

The critical technology of the software reuse is component technology. The critical technology of the software reuse is component technology, the most basic and important technology of software component is the component extract technology. Most of the existing component extraction method is done after the abstract of UML diagram. But on the whole the existing components extraction method that based on the reverse engineering is quite cumbersome.This paper address these shortcomings, take full advantage of the advantages of UCM (Use Case Maps), combined with the expression method of UCM, UCM semantics and tools, the procedure of component extraction, the knowledge of component reusable measurement, etc . Then proposed the Measurement of reusability based on the stub and interface of UCM. The measurement method consists of three metrics: include independence, functional independence, temporary dependence. This simple and efficient measurement takes advantage of the benefits of UCM.Based on the analysis of UCM demand-dependent, use the reusability measurement method which used for component recognition and extraction of the legacy system, this paper gives a specific component extraction algorithm CEABOU algorithm. The algorithm first identifies the core stubs, then these stubs as the starting point for the expansion of access to heritage components. In the method,the core stubs are firstly identified,and then spread is performed from these stubs to get the legacy component. The simple and efficient extraction method prevent the extraction of reverse engineering, so the method is more reasonable.Then the component extraction model is designed based on the algorithm, the component extraction model has been applied to online marketing system, combined with the quality evaluation of components and reusability verification theory to evaluate and verify the quality of the results that extracted from the legacy system, confirmed the effectiveness of component extraction method.