Research On Service Selection Method Based On Combination Auction

With the proliferation of cloud computing and "software as a service concept",the main forms,operating modes,production methods and usage patterns of software systems in the Internet environment are undergoing tremendous changes.Recently,with the explosive growth of the number of services,a large number of services with the same functions and different non-functional characteristics are distributed on the network.How to select a service that meets the needs of users with better quality and reliable operation among a large set of functionally equivalent services becomes an urgent problem to be solved.In this competitive environment,the selection of services is a complex decision process involving many stakeholders.The benefits obtained by service providers and candidate service providers are inextricably linked,and their needs cannot simultaneously Get the most satisfaction.And each candidate service provider is a competitive relationship and a part of mutual cooperation.Therefore,combined auction as a mode to effectively solve the problem of multi-quality service selection is more and more applied to the process of SBS service selection.However,most of the current methods of using combined auction to solve service selection only copy the combined auction model directly to the service selection problem.This approach ignores the characteristics of the web service selection problem,resulting in the quality and efficiency of service selection.The degree of improvement.Therefore,this dissertation proposes an in-depth study of service selection based on the optimization method of combined auction.The main research is reflected in the following aspects:(1)A method of selecting and optimizing the combination auction service based on relevance perception is proposed.The idea is to improve the overall performance of SBS by obtaining the potential association relationship between candidate services.In view of the above ideas,the service association relationship is defined,and the extraction framework and method of the service association relationship are given.At the same time,a multi-attribute combination auction service selection method based on association relationship is proposed to express the relationship between candidate services,so as to select suitable candidate services more efficiently.On this basis,the system describes the auction process of combined auctions considering the relationship between services,and proposes a service selection model and support mechanism based on service association.(2)The quality of acquisition for demand is the key to the success of SBS.The effort to check and determine the missing or wrong demand during the demand collection and acquisition phase is huge,time consuming and error prone,and the design of the inquiry mechanism is The core of the combined auction determines the efficiency and effectiveness of the combined auction.This chapter proposes a demand-oriented adaptive inquiry combination auction service selection method,which uses the existing similar SBS domain knowledge,ie SBS feature model,to guide the user to describe the requirements,and uses the recommendation system as the SBS to be developed with partial features.Recommended features for demand-assisted acquisition.At the same time,the QoS attribute weight of SBS demand is determined by AHP,and the demand-oriented service selection model is determined.At the same time,this chapter proposes a self-adaptive inquiry model,which uses the minimum attenuation rate to more accurately perform adaptive inquiry based on the historical bid information of the candidate service providers,thereby improving the efficiency and quality of the combined auction.(3)With the continuous improvement of the requirements of SBS designers,the solution model of the competitive bidding problem is often modeled as a multi-objective problem.As the complexity of the problem is gradually improved,the use of intelligent algorithms to solve the competitive bid problem can improve the solution.Quality and efficiency.An modified multi-objective discrete artificial bee colony algorithm MMOABC for solving the problem of selecting competitive bidding for combined auction service is proposed.The algorithm is a strategy and parameter adaptive adjustment algorithm for multi-objective complex problems.No matter how complicated the requirements of SBS designers can be,the overall performance of SBS can be maximized,and the cost is minimized.To the most suitable candidate service combination.(4)The cloud service presents a large number of features,divergence,dynamic evolution,and heterogeneity.A multi-agent cloud service selection method based on tree structure combined auction is proposed.This method is a service selection optimization method that considers the characteristics of cloud services,which can promote SBS designers and candidate cloud service providers in their respective revenue and service QoS.A trade-off between requirements,this method not only ensures that the services provided by the candidate cloud service providers meet the requirements of the SBS designer SLA,but also achieves the relative maximization of the respective benefits of the candidate service providers and SBS designers,and is efficient.The tree structure combined auction constraint is added to reasonably limit the target submitted by the candidate service provider,which reduces the computational complexity of solving the winning bid problem in the cloud computing environment and improves the efficiency of cloud service selection.