Analyzing the Effectiveness of the Gaussian Mixture Model Clustering Algorithm in Software Enhancement Effort Estimation

Kód výsledku v IS VaVaI

<a href="https://www.isvavai.cz/riv?ss=detail&h=RIV%2F70883521%3A28140%2F22%3A63556598" target="_blank" >RIV/70883521:28140/22:63556598 - isvavai.cz</a>

Výsledek na webu

<a href="https://link.springer.com/chapter/10.1007/978-3-031-21967-2_21" target="_blank" >https://link.springer.com/chapter/10.1007/978-3-031-21967-2_21</a>

DOI - Digital Object Identifier

<a href="http://dx.doi.org/10.1007/978-3-031-21967-2_21" target="_blank" >10.1007/978-3-031-21967-2_21</a>

Jazyk výsledku

angličtina

Název v původním jazyce

Analyzing the Effectiveness of the Gaussian Mixture Model Clustering Algorithm in Software Enhancement Effort Estimation

Popis výsledku v původním jazyce

Background: The influence of data clustering on the effort estimating process has been studied extensively. Studies focus on partitioning and density-based clustering, and some use hierarchical clustering, but most focus on software development effort estimation. Aim: We focus on the Gaussian Mixture Model algorithm’s effectiveness in the software enhancement effort estimation. Method: We used the Gaussian Mixture Model clustering algorithm to cluster the dataset into clusters and then applied the IFPUG FPA method for effort estimation on these clusters. The ISBSG dataset was used in this study. The number of clusters is determined using the Elbow method with the Distortion score. Besides, the k-means algorithm was also used as the comparative algorithm. The baseline model was determined by using the FPA method on the entire dataset without clustering. Result: With the number of clusters selected as 4, on six evaluation criteria, MAE, MAPE, RMSE, MBRE, and MIBRE, the experimental results show the estimated accuracy using the FPA method on clustered data significantly better when compared with no clustering. Conclusion: the software enhancement effort estimation can be significantly improved when using the Gaussian Mixture Model clustering algorithm

Název v anglickém jazyce

Analyzing the Effectiveness of the Gaussian Mixture Model Clustering Algorithm in Software Enhancement Effort Estimation

Popis výsledku anglicky

Background: The influence of data clustering on the effort estimating process has been studied extensively. Studies focus on partitioning and density-based clustering, and some use hierarchical clustering, but most focus on software development effort estimation. Aim: We focus on the Gaussian Mixture Model algorithm’s effectiveness in the software enhancement effort estimation. Method: We used the Gaussian Mixture Model clustering algorithm to cluster the dataset into clusters and then applied the IFPUG FPA method for effort estimation on these clusters. The ISBSG dataset was used in this study. The number of clusters is determined using the Elbow method with the Distortion score. Besides, the k-means algorithm was also used as the comparative algorithm. The baseline model was determined by using the FPA method on the entire dataset without clustering. Result: With the number of clusters selected as 4, on six evaluation criteria, MAE, MAPE, RMSE, MBRE, and MIBRE, the experimental results show the estimated accuracy using the FPA method on clustered data significantly better when compared with no clustering. Conclusion: the software enhancement effort estimation can be significantly improved when using the Gaussian Mixture Model clustering algorithm

Druh

D - Stať ve sborníku

CEP obor

—

OECD FORD obor

10201 - Computer sciences, information science, bioinformathics (hardware development to be 2.2, social aspect to be 5.8)

Projekt

—

Návaznosti

S - Specificky vyzkum na vysokych skolach<br>I - Institucionalni podpora na dlouhodoby koncepcni rozvoj vyzkumne organizace

Rok uplatnění

2022

Kód důvěrnosti údajů

S - Úplné a pravdivé údaje o projektu nepodléhají ochraně podle zvláštních právních předpisů

Název statě ve sborníku

Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics)

ISBN

978-3-031-21966-5

ISSN

0302-9743

e-ISSN

—

Počet stran výsledku

14

Strana od-do

255-268

Název nakladatele

Springer Science and Business Media Deutschland GmbH

Místo vydání

Berlín

Místo konání akce

Ho Chi Minh City

Datum konání akce

28. 11. 2022

Typ akce podle státní příslušnosti

WRD - Celosvětová akce

Kód UT WoS článku

000916496900021