Please use this identifier to cite or link to this item: http://hdl.handle.net/10397/12214
Title: Kernel methods for software effort estimation effects of different kernel functions and bandwidths on estimation accuracy
Authors: Kocaguneli, E
Menzies, T
Keung, JW
Keywords: Effort estimation
Data mining
Kernel function
Bandwidth
Issue Date: 2013
Publisher: Springer
Source: Empirical software engineering, 2013, v. 18, no. 1, p. 1-24 How to cite?
Journal: Empirical Software Engineering 
Abstract: Analogy based estimation (ABE) generates an effort estimate for a new software project through adaptation of similar past projects (a.k.a. analogies). Majority of ABE methods follow uniform weighting in adaptation procedure. In this research we investigated non-uniform weighting through kernel density estimation. After an extensive experimentation of 19 datasets, 3 evaluation criteria, 5 kernels, 5 bandwidth values and a total of 2090 ABE variants, we found that: (1) non-uniform weighting through kernel methods cannot outperform uniform weighting ABE and (2) kernel type and bandwidth parameters do not produce a definite effect on estimation performance. In summary simple ABE approaches are able to perform better than much more complex approaches. Hence,-provided that similar experimental settings are adopted-we discourage the use of kernel methods as a weighting strategy in ABE.
URI: http://hdl.handle.net/10397/12214
ISSN: 1382-3256
DOI: 10.1007/s10664-011-9189-1
Appears in Collections:Journal/Magazine Article

Access
View full-text via PolyU eLinks SFX Query
Show full item record

SCOPUSTM   
Citations

23
Last Week
0
Last month
1
Citations as of Aug 16, 2018

WEB OF SCIENCETM
Citations

17
Last Week
0
Last month
0
Citations as of Aug 17, 2018

Page view(s)

58
Last Week
0
Last month
Citations as of Aug 13, 2018

Google ScholarTM

Check

Altmetric


Items in DSpace are protected by copyright, with all rights reserved, unless otherwise indicated.