The Research And Implementation Of Reusable Software Asset Management

With the enlarging of software scale,the growth of software complexity and the exponential increment of code,software developers face many problems including increasement of software development cost,delaying of the development cycle and lower quality of software product.Thus,software reuse is one of the most effective approaches to solve those problems.This technique reduces time of software development cycle,lowers the cost of production,and increases the production efficiency through organizing reusable software components in the form of component repository.However,previous simple reuse technique recently can not satisfy enterprise users’ growing demand, which mainly due to the following three aspects:(1)The description of component does not have a unified standard in component repository,which leads to the difficulty in exchanging and sharing of variety applications.(2)Having single type,component can not be extended.What’s more;it can’t effectively support software reuse during each stages of software development life cycle. (3)Traditional software retrieval mainly adopts the method based on keyword matching or the content classified directory technique, lacking of intelligent processing.Ambiguity of a word as well as misunderstanding of semantic information results in the lower precision and recall rate.As to the problems above,this paper introduces a new concept of software asset.Firstly,on the basis in Reusable Asset Specification proposed by OMG,this thesis describes software asset from some aspects, such as basic information,classified information,interface Specification, etc.After that,a set of XML-based,suitable for software resue,united and standard description model of reusable asset is constructed. Secondly,using the advantage of ontology which describes domain concept and relationship explicitly.This paper introduces semantic into asset description model,and builds a ontology model of asset coding in OWL.And then,it devises a semantic information automatic extraction algorithm(XMO)including mapping between XML and OWL, researches on the use of Prot(?)g(?)to construct asset ontology as well.What’s more,a rational storage strategy is designed.Based on the advantages and disadvantages of exsiting search mechanisms,an ontology-based semantic rewrite algorithm is proposed, in order to extend the user’s searching request semantically.A new model is designed for calculating semantic similarity.According to the algorithm and calculating model,this paper gives a semantic retrieval scheme and an implementation of reusable software asset.Afterwards, with the purpose of reusing enterprise software asset,a resue evaluation model is constructed.At last,a software asset management prototype system based on Struts Framework is also realized.This system contains the function modules of asset adding,asset searching and background managing.Experiments are shown that the new approach has a good performance on the recall and precision rate of asset retrieval,which provides a feasible solution for enterprise asset management.This thesis refers to advanced technologies in software resue, ontology,information retrieval,etc.,with successfully carrying out the contruction and retrival processing of enterprise ontology-based reusable asset.The software asset management method which is extended to the semantic domain is a new attempt in this field,having far research significance and real application value.