The Design And Application Of Digital Asset Management System In Colleges And Universities

The digital asset,which has been an important part of the assets management in colleges,is a kind of fixed assets.But,its management style,which is more inclined to the classification of assets,is different from tangible assets such as academic Building,office equipment,books.In the existing digital asset management system,C-means clustering algorithm,in most instances,is used to deal with repetition rate of students' art works.But there are two problems in practice.The first one is that the result of duplicate checking for students' art works is non-convergence in the condition of different clustering initial points.And the other one is that the algorithm is easy to converge to the local optimal solution in the clustering process.Aimed at above issues,this paper designs and builds an entire digital asset management system which is designed for colleges.Adopting an improved clustering algorithm,this system,which is based on the need of business transformation for the digital library and the need of business process adjustments,is implemented.And this system is suitable for the Industry characteristics and business characteristics of asset management in Colleges.Basing on the theoretical principle of SOA(service-oriented architecture)and system construction method,this paper,by using web services,achieve an asset management system on the server and PC terminal.Being based on the J2 EE framework mainly and using the idea of layered software development,this paper designs and implements a kind of special digital asset management system.And this system,which adopts the mixed B/S and C/S frame,improves the efficiency and stability above.The system has completed many functions,which contains user management,audio and video digital asset management,picture digital asset management,image digital asset filter,correlation statistics and so on.In addition,aiming at the problems,which are that the repetition rate result of students' art works are unstably and the clustering process convergences to local optimal solution,this paper,by introducing the theory of genetic algorithm,improves a new algorithm which is based on dual genetic algorithm.The improved C-means clustering algorithm in this paper has the following advantages: on the one hand the first genetic algorithm is used to optimize the cluster center,and the number of the genetic center is determined by the analysis of the individual,on the other hand the individual is the center of a cluster,the double genetic algorithm fitness,so as to determine the global optimal solution of the clustering center.Under the conditions of master map,grayscale open operation,gray closed operation and so on,the results show that the improved clustering algorithm is more effective than the Kmeans and C-means algorithm.