software defect prediction using fuzzy support vector regression

ISCAS OpenIR > 2010软件所会议论文

	software defect prediction using fuzzy support vector regression
	Yan Zhen; Chen Xinyu; Guo Ping
	2010
会议名称	7th International Symposium on Neural Networks, ISNN 2010
会议录名称	Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics)
页码	17-24
会议日期	43988
会议地点	Shanghai, China
收录类别	ei
出版地	Germany
ISSN	3029743
ISBN	3642133177
部门归属	(1) School of Computer, Beijing Institute of Technology, Beijing 100081, China; (2) State Key Laboratory of Computer Science, Institute of Software, Chinese Academy of Sciences, Beijing 100190, China
摘要	Regression techniques have been applied to improve software quality by using software metrics to predict defect numbers in software modules. This can help developers allocate limited developing resources to modules containing more defects. In this paper, we propose a novel method of using Fuzzy Support Vector Regression (FSVR) in predicting software defect numbers. Fuzzification input of regressor can handle unbalanced software metrics dataset. Compared with the approach of support vector regression, the experiment results with the MIS and RSDIMU datasets indicate that FSVR can get lower mean squared error and higher accuracy of total number of defects for modules containing large number of defects. © 2010 Springer-Verlag.
关键词	Computer Software Selection And Evaluation Defects Forecasting Regression Analysis Vectors
主办者	Shanghai Jiao Tong University; The Chinese University of Hong Kong; IEEE Shanghai Section; International Neural Network Society; IEEE Computational Intelligence Society
语种	英语
内容类型	会议论文
URI标识	http://ir.iscas.ac.cn/handle/311060/8906
专题	2010软件所会议论文
推荐引用方式 GB/T 7714	Yan Zhen,Chen Xinyu,Guo Ping. software defect prediction using fuzzy support vector regression[C]. Germany,2010:17-24.

条目包含的文件
文件名称/大小	文献类型	版本类型	开放类型	使用许可
software defect pred（131KB）			限制开放	--	请求全文