Hyper-parameter tuning of classification and regression trees for software effort estimation

 

Αποθηκεύτηκε σε:
Λεπτομέρειες βιβλιογραφικής εγγραφής
Συγγραφείς: Villalobos Arias, Leonardo, Quesada López, Christian Ulises, Martínez Porras, Alexandra, Jenkins Coronas, Marcelo
Μορφή: comunicación de congreso
Ημερομηνία έκδοσης:2021
Περιγραφή:Classification and regression trees (CART) have been reported to be competitive machine learning algorithms for software effort estimation. In this work, we analyze the impact of hyper-parameter tuning on the accuracy and stability of CART using the grid search, random search, and DODGE approaches. We compared the results of CART with support vector regression (SVR) and ridge regression (RR) models. Results show that tuning improves the performance of CART models up to a maximum of 0.153 standardized accuracy and reduce its stability radio to a minimum of 0.819. Also, CART proved to be as competitive as SVR and outperformed RR.
Χώρα:Kérwá
Ίδρυμα:Universidad de Costa Rica
Repositorio:Kérwá
Γλώσσα:Inglés
OAI Identifier:oai:kerwa.ucr.ac.cr:10669/102209
Διαθέσιμο Online:https://hdl.handle.net/10669/102209
https://doi.org/10.1007/978-3-030-72660-7_56
Λέξη-Κλειδί :software effort estimation
hyper-parameter tuning