Research On Design Pattern Selection Model Based On CBR

Design patterns are empirical documents used to solve recurring software design problems in a specific context.Since the experience and knowledge encapsulated by design patterns are more professional and the organizational forms are different,as design patterns continue to be proposed,their expressed intentions and scope of application become more and more complex.For specific software design problems,it is very challenging for software developers to choose the right design pattern to solve these problems without the support of tools.The existing software design pattern selection methods mainly have the following problems: lack of flexibility when the number of design patterns is large,insufficient utilization of the search scenarios of design patterns,and the need for more manual participation.This paper proposes a case-based reasoning(CBR)method to solve the problem of design pattern selection.Because the core idea of case-based reasoning and design patterns is to use the successful experience of solving problems in the past to solve emerging problems,the internal connection between the two makes it possible to use casebased reasoning technology to solve the design pattern selection problem.In case-based reasoning,the increase in the number of cases can improve the performance of the casebased reasoning model to a certain extent,so it can solve the problem that the increase in the number of design patterns of existing methods affects flexibility.The genetic algorithm is introduced to optimize the parameters,which solves the problem that the model depends on manual participation.This article first introduces the relationship between design patterns and case-based reasoning,and reviews the advantages and disadvantages of the existing design pattern selection methods;secondly,using the inherent connection between design patterns and case-based reasoning ideas,a design pattern based on case-based reasoning is proposed The model was selected and the case representation of the design pattern was designed,including the representation of the design pattern as a case,the determination of similar cases,and the maintenance strategy of the case library;then a strategy for adding cases was designed to solve the existing case library.There are cases where the query requirements are not met,and an intent-based ordering method of design patterns is proposed,the calculation method of related similarity is defined and the corresponding proof is given;again,a parameter adjustment based on genetic algorithm is proposed.The optimal method is used to solve the problem that the parameters defined in the similarity calculation method are excessively dependent on the designer's experience,and the corresponding fitness function,genetic representation and coding components are designed to adaptively adjust the parameters involved in the similarity calculation Finally,the method proposed in this paper is applied to the actual data set and design pattern set in software development,and verified and analyzed.The experimental results show that the proposed method has better matching rate and sorting accuracy than other methods.There is a certain improvement.