摘要
随着软件系统的规模和复杂性日益增长,软件开发已经演变成一项复杂的系统工程。软件工程中的对象、活动和过程更加难以控制和管理,因此该领域原有的经验直觉型的处理模式已经不能适应新的需求,而数据挖掘技术的引入为实现知识智能型软件工程提供了重要契机。以软件工程领域中的数据对象为主线,对在程序代码分析、故障检测、软件项目管理、开源软件开发等软件活动中所运用到的数据挖掘技术进行了系统的介绍和归纳,并在每一环节作了方法间的优劣性对比分析。最后还指出了若干值得进一步研究的方向。
With rapid increase of the size and complexity of software system, software development has evolved into a complex and systematic engineering. The objects, activities and processes in software engineering will to be controlled even more with difficulty, so the traditional modes, such as treatment with experience and intuition, are unable to adapt properly to the new requirements. However, the introduction of data mining techniques can promote the development of knowledgeable and intelligent software engineering. In the perspective of the data to be mined in software engineering field,the paper systematically described and summarized the data mining techniques adopted in the activities such as program code analysis, fault detection, software project management, and open source software （OSS） development. The comparison analysis about these techniques is also addressed in each section. Furthermore, some on-going research issues in this direction are also discussed in the end.
出处
《计算机科学》
CSCD
北大核心
2009年第5期1-6,26,共7页
Computer Science
基金
国家自然科学基金项目(60803046,70571025)
中国博士后科学基金(20070410946)
湖北省自然科学基金(2005ABA266)
江西省教育厅科技项目(赣教技字[2007]267号)资助
关键词
数据挖掘
软件工程
预测
软件项目管理
开源软件
Data mining, Software engineering, Prediction, Software project management, OSS
