Abstract
The performance of software cost estimation based on analogy reasoning depends upon the measures that specifying the similarity between software projects. This paper empirically investigates the use of probabilistic-based distance functions to improve the similarity measurement. The probabilistic-based distance functions are considerably more robust, because they collect the implicit correlation between the occurrences of project feature attributes. This information gain enables the constructed estimation model to be more concise and comprehensible. The study compares 6 probabilistic-based distance functions against the commonlyused Euclidian distance. We empirically evaluate the implemented cost estimation model using 5 real-world datasets collected from the PROMISE repository. The result shows a significant improvement in terms of error reduction, that implies an estimation based on probabilistic-based distance functions achieve higher accuracy on average, and the peak performance significantly outperforms the Euclidian distance based on Wilcoxon signed-rank test.
| Original language | English |
|---|---|
| Title of host publication | APSEC 2013 - Proceedings of the 20th Asia-Pacific Software Engineering Conference |
| Publisher | IEEE Computer Society |
| Pages | 541-546 |
| Number of pages | 6 |
| Volume | 1 |
| ISBN (Print) | 9781479921430, 9780769549224 |
| DOIs | |
| Publication status | Published - 2013 |
| Externally published | Yes |
| Event | 20th Asia-Pacific Software Engineering Conference, APSEC 2013 - Bangkok, Thailand Duration: Dec 2 2013 → Dec 5 2013 |
Other
| Other | 20th Asia-Pacific Software Engineering Conference, APSEC 2013 |
|---|---|
| Country | Thailand |
| City | Bangkok |
| Period | 12/2/13 → 12/5/13 |
Fingerprint
Keywords
- Analogy
- Heterogeneous distance function
- Machine learning
- Software cost estimation
ASJC Scopus subject areas
- Software
Cite this
Improving analogy-based software cost estimation through probabilistic-based similarity measures. / Phannachitta, Passakorn; Keung, Jacky; Monden, Akito; Matsumoto, Ken Ichi.
APSEC 2013 - Proceedings of the 20th Asia-Pacific Software Engineering Conference. Vol. 1 IEEE Computer Society, 2013. p. 541-546 6805449.Research output: Chapter in Book/Report/Conference proceeding › Conference contribution
}
TY - GEN
T1 - Improving analogy-based software cost estimation through probabilistic-based similarity measures
AU - Phannachitta, Passakorn
AU - Keung, Jacky
AU - Monden, Akito
AU - Matsumoto, Ken Ichi
PY - 2013
Y1 - 2013
N2 - The performance of software cost estimation based on analogy reasoning depends upon the measures that specifying the similarity between software projects. This paper empirically investigates the use of probabilistic-based distance functions to improve the similarity measurement. The probabilistic-based distance functions are considerably more robust, because they collect the implicit correlation between the occurrences of project feature attributes. This information gain enables the constructed estimation model to be more concise and comprehensible. The study compares 6 probabilistic-based distance functions against the commonlyused Euclidian distance. We empirically evaluate the implemented cost estimation model using 5 real-world datasets collected from the PROMISE repository. The result shows a significant improvement in terms of error reduction, that implies an estimation based on probabilistic-based distance functions achieve higher accuracy on average, and the peak performance significantly outperforms the Euclidian distance based on Wilcoxon signed-rank test.
AB - The performance of software cost estimation based on analogy reasoning depends upon the measures that specifying the similarity between software projects. This paper empirically investigates the use of probabilistic-based distance functions to improve the similarity measurement. The probabilistic-based distance functions are considerably more robust, because they collect the implicit correlation between the occurrences of project feature attributes. This information gain enables the constructed estimation model to be more concise and comprehensible. The study compares 6 probabilistic-based distance functions against the commonlyused Euclidian distance. We empirically evaluate the implemented cost estimation model using 5 real-world datasets collected from the PROMISE repository. The result shows a significant improvement in terms of error reduction, that implies an estimation based on probabilistic-based distance functions achieve higher accuracy on average, and the peak performance significantly outperforms the Euclidian distance based on Wilcoxon signed-rank test.
KW - Analogy
KW - Heterogeneous distance function
KW - Machine learning
KW - Software cost estimation
UR - http://www.scopus.com/inward/record.url?scp=84936889065&partnerID=8YFLogxK
UR - http://www.scopus.com/inward/citedby.url?scp=84936889065&partnerID=8YFLogxK
U2 - 10.1109/APSEC.2013.78
DO - 10.1109/APSEC.2013.78
M3 - Conference contribution
AN - SCOPUS:84936889065
SN - 9781479921430
SN - 9780769549224
VL - 1
SP - 541
EP - 546
BT - APSEC 2013 - Proceedings of the 20th Asia-Pacific Software Engineering Conference
PB - IEEE Computer Society
ER -