首页 | 本学科首页   官方微博 | 高级检索  
     

结合扇入分析和聚类分析的Aspect挖掘方法
引用本文:葛君伟,张鹏,方义秋. 结合扇入分析和聚类分析的Aspect挖掘方法[J]. 计算机应用研究, 2009, 26(11): 4128-4132. DOI: 10.3969/j.issn.1001-3695.2009.11.037
作者姓名:葛君伟  张鹏  方义秋
作者单位:重庆邮电大学,计算机学院,重庆,400065;重庆邮电大学,计算机学院,重庆,400065;重庆邮电大学,计算机学院,重庆,400065
基金项目:重庆市自然科学基金资助项目(2005BB2059);重庆市教委科学技术项目(KJ040511)
摘    要:结合扇入分析和聚类分析提出一种新的Aspect挖掘方法,以方法的频繁调用作为横切关注点的外在特征,首先对目标系统进行扇入分析,获取扇入值符合条件的方法集合和调用方法集合,把调用方法看作这些方法的特征属性,构造对象数据矩阵进行聚类分析获得候选Aspect种子集。该方法在保证较高召回率的同时,进一步提高了挖掘的准确率和执行效率。

关 键 词:面向方面编程  横切关注点  方面挖掘  扇入分析  聚类分析

New Aspect mining approach based on fan-in analysis and clustering analysis
GE Jun-wei,ZHANG Peng,FANG Yi-qiu. New Aspect mining approach based on fan-in analysis and clustering analysis[J]. Application Research of Computers, 2009, 26(11): 4128-4132. DOI: 10.3969/j.issn.1001-3695.2009.11.037
Authors:GE Jun-wei  ZHANG Peng  FANG Yi-qiu
Affiliation:(College of Computer Science & Technology, Chongqing University of Post & Telecommunications, Chongqing 400065, China)
Abstract:This paper proposed a new approach of aspect mining based on fan-in and clustering analysis. This approach regard frequented called functions as the character of a crosscutting concern. Firstly, this paper analyzed the aimed system through the fan-in theory, got the called functions whose fan-in value according to be request. Then, did the clustering according the matrix whose objects were called functions and the attributes were the called functions, getting the candidate aspect seeds. This approach had raised the precision and efficiency of the aspect mining, while also kept a high recall.
Keywords:aspect-oriented programming(AOP)   crosscutting concerns   aspect mining   fan-in analysis   clustering analysis
本文献已被 万方数据 等数据库收录!
点击此处可从《计算机应用研究》浏览原始摘要信息
点击此处可从《计算机应用研究》下载全文
设为首页 | 免责声明 | 关于勤云 | 加入收藏

Copyright©北京勤云科技发展有限公司  京ICP备09084417号