on the predictability of software efforts using machine learning techniques

ISCAS OpenIR

	on the predictability of software efforts using machine learning techniques
	Zhang Wen; Yang Ye; Wang Qing
	2011
会议名称	6th International Conference on Evaluation of Novel Approaches to Software Engineering, ENASE 2011
会议录名称	ENASE 2011 - Proceedings of the 6th International Conference on Evaluation of Novel Approaches to Software Engineering
页码	5-14
会议日期	June 8, 2011 - June 11, 2011
会议地点	Beijing, China
收录类别	EI
ISBN	9789898425577
部门归属	(1) Laboratory for Internet Software Technologies Institute of Software Chinese Academy of Sciences Beijing 100190 China
摘要	This paper investigates the predictability of software effort using machine learning techniques. We employed unsupervised learning as k-medoids clustering with different similarity measures to extract natural clusters of projects from software effort data set, and supervised learning as J48 decision tree, back propagation neural network (BPNN) and na¨ive Bayes to classify the software projects. We also investigate the impact of imputing missing values of projects on the performances of both unsupervised and supervised learning techniques. Experiments on ISBSG and CSBSG data sets demonstrate that unsupervised learning as k-medoids clustering has produced a poor performance in software effort prediction and Kulzinsky coefficient has the best performance in software effort prediction in measuring the similarities of projects. Supervised learning techniques have produced superior performances in software effort prediction. Among the three supervised learning techniques, BPNN produces the best performance. Missing data imputation has improved the performances of both unsupervised and supervised learning techniques.; This paper investigates the predictability of software effort using machine learning techniques. We employed unsupervised learning as k-medoids clustering with different similarity measures to extract natural clusters of projects from software effort data set, and supervised learning as J48 decision tree, back propagation neural network (BPNN) and na¨ive Bayes to classify the software projects. We also investigate the impact of imputing missing values of projects on the performances of both unsupervised and supervised learning techniques. Experiments on ISBSG and CSBSG data sets demonstrate that unsupervised learning as k-medoids clustering has produced a poor performance in software effort prediction and Kulzinsky coefficient has the best performance in software effort prediction in measuring the similarities of projects. Supervised learning techniques have produced superior performances in software effort prediction. Among the three supervised learning techniques, BPNN produces the best performance. Missing data imputation has improved the performances of both unsupervised and supervised learning techniques.
关键词	Decision Trees Forecasting Neural Networks Software Engineering Supervised Learning Unsupervised Learning
语种	英语
内容类型	会议论文
URI标识	http://ir.iscas.ac.cn/handle/311060/16307
专题	中国科学院软件研究所
推荐引用方式 GB/T 7714	Zhang Wen,Yang Ye,Wang Qing. on the predictability of software efforts using machine learning techniques[C],2011:5-14.