A study on software effort prediction using machine learning techniques

ISCAS OpenIR

	A study on software effort prediction using machine learning techniques
	Zhang, Wen (1); Yang, Ye (1); Wang, Qing (1)
	2013
会议名称	6th International Conference Evaluation of Novel Approaches to Software Engineering, ENASE 2011
页码	1-15
会议日期	June 8, 2011 - June 11, 2011
会议地点	Beijing, China
收录类别	EI
出版地	Springer Verlag, Tiergartenstrasse 17, Heidelberg, D-69121, Germany
ISSN	18650929
ISBN	9783642323409
部门归属	(1) Laboratory for Internet Software Technologies, Institute of Software, Chinese Academy of Sciences, Beijing 100190, China
摘要	This paper conducts a study on of software effort prediction using machine learning techniques. Both supervised and unsupervised learning techniques are employed to predict software effort using historical dataset. The unsupervised learning as k-medoids clustering equipped with different similarity measures is used to cluster projects in historical dataset. The supervised learning as J48 decision tree, back propagation neural network (BPNN) and na¨ive Bayes is used to classify the software projects into different effort classes. We also impute the missing values in the historical datasets and then machine learning techniques are adopted to predict software effort. Experiments on ISBSG and CSBSG datasets demonstrate that unsupervised learning as k-medoids clustering produced a poor performance. Kulzinsky coefficient has the best performance in measuring the similarities of projects. Supervised learning techniques produced superior performances than unsupervised learning techniques in software effort prediction. BPNN produced the best performance among the three supervised learning techniques. Missing data imputation improved the performances of both unsupervised and supervised learning techniques in software effort prediction. © Springer-Verlag Berlin Heidelberg 2013.; This paper conducts a study on of software effort prediction using machine learning techniques. Both supervised and unsupervised learning techniques are employed to predict software effort using historical dataset. The unsupervised learning as k-medoids clustering equipped with different similarity measures is used to cluster projects in historical dataset. The supervised learning as J48 decision tree, back propagation neural network (BPNN) and na¨ive Bayes is used to classify the software projects into different effort classes. We also impute the missing values in the historical datasets and then machine learning techniques are adopted to predict software effort. Experiments on ISBSG and CSBSG datasets demonstrate that unsupervised learning as k-medoids clustering produced a poor performance. Kulzinsky coefficient has the best performance in measuring the similarities of projects. Supervised learning techniques produced superior performances than unsupervised learning techniques in software effort prediction. BPNN produced the best performance among the three supervised learning techniques. Missing data imputation improved the performances of both unsupervised and supervised learning techniques in software effort prediction. © Springer-Verlag Berlin Heidelberg 2013.
语种	英语
内容类型	会议论文
URI标识	http://ir.iscas.ac.cn/handle/311060/16675
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 . A study on software effort prediction using machine learning techniques[C]. Springer Verlag, Tiergartenstrasse 17, Heidelberg, D-69121, Germany,2013:1-15.