Mõningatest tarkvara vigade hindamise mudelitest
Abstract
Käesoleva töö peamiseks eesmärgiks on anda üldisem ülevaade protsessidest tarkvara vigade hindamise mudelites, mis kasutavad masinõppe klassifikaatoreid, ja analüüsida mõningaid hindamiseskperimentide tulemusi, mis on läbi viidud antud töös refereeritud uurimistöödes. Lisaks on antud lühike selgitus antud töös vaadeldavates tarkvara vigade hindamise mudelites kasutatud algoritmidest ja tuuakse välja ning seletatakse lahti mõned hinnangumõõdikud, mida kasutatakse tarkvara vigade hindamise mudelite hindamistäpsuste mõõtmiseks. Tuuakse välja ka üldine ülevaade vaadeldavates tarkvara vigade hindamise mudelites toimuvatest protsessidest.
The main idea of this thesis is to give a general overview of the processes within the soft-ware defect prediction models using machine learning classifiers and to provide analysis to some of the results of the evaluation experiments conducted in the research papers covered in this work. Additionally, a brief explanation of the algorithms used within the software defect prediction models covered in this work is given and some of the evaluation measures used to evaluate the prediction accuracy of software defect prediction models are listed and explained. Also, a general overview of the processes within a handful of specific software defect prediction models is provided.
The main idea of this thesis is to give a general overview of the processes within the soft-ware defect prediction models using machine learning classifiers and to provide analysis to some of the results of the evaluation experiments conducted in the research papers covered in this work. Additionally, a brief explanation of the algorithms used within the software defect prediction models covered in this work is given and some of the evaluation measures used to evaluate the prediction accuracy of software defect prediction models are listed and explained. Also, a general overview of the processes within a handful of specific software defect prediction models is provided.