摘要
提出了将数据挖掘应用于软件工程学中,数据挖掘是一个涉及多领域的交叉学科,它拥有许多成熟的技术。其中,模糊聚类法就是数据挖掘的核心技术之一。在对大量的软件工程数据进行挖掘处理后,经过λ水平截集,将相似的模块聚为一类,作为模块评估的基础,从中得出有价值的知识模式,用于指导软件工程实践。
This paper puts forward applying data mining to software engineering. Data mining is an intersection subject involving many fields, and owns many mature technologies. Fuzzy clustering is often used in data mining. After handling a large quantity of software data, the paper clusters the similar module and makes it become the base of module estimation. It can identify some valuable patterns, which can be used to direct software engineering practice. Compared with the traditional method, the data mining method is more direct, more efficient and easier to realize.
出处
《计算机工程》
EI
CAS
CSCD
北大核心
2005年第8期52-53,139,共3页
Computer Engineering
基金
国家自然科学基金资助项目(60075016)
关键词
软件工程
数据挖掘
聚类
机器学习
模块评估
相似关系
Software engineering
Data mining
Clustering
Machine leaning
Module estimation
Similar relation