设计模式识别的特征信息分类研究  

FEATURE INFORMATION CLASSIFICATION OF DESIGN PATTERN RECOGNITION

作　　者：肖卓宇 何锫[2] 徐运标 陈果[1] 郭杰[1] 王林春 Xiao Zhuoyu;He Pei;Xu Yunbiao;Chen Guo;Guo Jie;Wang Linchun(School of Information Engineering,Hunan Industry Polytechnic,Changsha 410208,Hunan,China;School of Computer Science&Education Software,Guangzhou University,Guangzhou 510006,Guangdong,China)

机构地区：[1]湖南工业职业技术学院信息工程学院,湖南长沙410208 [2]广州大学计算机科学与教育软件学院,广东广州510006

出　　处：《计算机应用与软件》2022年第6期7-11,20,共6页Computer Applications and Software

基　　金：湖南省自然科学基金科教联合项目(2020JJ7035);湖南省教育厅科学研究项目(19C0611);国家自然科学基金项目(61977018)。

摘　　要：为设计模式识别精确率不高的问题,提出一种基于特征分类表示的设计模式识别方法,分结构型、行为型及创建型三类归纳特征信息,三类模式分别以Adapter模式、Command模式、Factory Method模式为例,通过4个经典的测试系统进行设计模式识别实验,识别精确率分别96.8%、81.95%、42.85%,取得了较好的效果。To solve the low accuracy of design pattern recognition,this paper proposes a method for design pattern recognition based on feature classification representation.The characteristic information was summarized into three types:structural pattern,behavioral pattern and creative pattern.Taking adapter mode,command mode and factory method mode as examples,the design pattern recognition experiments were carried out through four classic test systems.The results show that their recognition accuracy rates are 96.8%,81.95%and 42.85%respectively,which has better performance.

关 键 词：设计模式 设计模式识别 分类 特征信息 

分 类 号：TP311[自动化与计算机技术—计算机软件与理论]