In this paper we study a bin-based estimation method of the amount of effort associated with code development. We investigate the following 3 variants to define the bins: (1) the same amount of data in a bin (SVM same #), (2) the same range for each bin (SVM same range) and (3) the bins made by Ward’s method (SVM Ward). We carry out evaluation experiments to compare the accuracy of the proposed SVM models with that of the ε-SVR using Welch’s t-test and effect sizes. These results indicate that the methods SVM same # (1) and SVM Ward (3) can improve the accuracy of estimating the amount of effort in terms of the mean percentage of predictions that fall within 25 % of the actual value.

PDF, HTML

In {C}omputer and {I}nformation {S}cience , Roger Lee (Eds.), Berlin 2016. Springer Verlag.

@incollection{ICIS2015-eladlieb, title={{B}in-{B}ased {E}stimation of the {A}mount of {E}ffort for {E}mbedded {S}oftware {D}evelopment {P}rojects with {S}upport {V}ector {M}achines}, author={Kazunori Iwata and Elad Liebman and Peter Stone and Toyoshiro Nakashima and Yoshiyuki Anan and Naohiro Ishii}, booktitle={{C}omputer and {I}nformation {S}cience }, editor={Roger Lee}, address={Berlin}, publisher={Springer Verlag}, url="http://www.cs.utexas.edu/users/ai-lab?liebman:icis15", year={2016} }