Making the Most of Small Software Engineering Datasets With Modern Machine Learning

Making the Most of Small Software Engineering Datasets With Modern Machine Learning

This paper provides a starting point for Software Engineering (SE) researchers and practitioners faced with the problem of training machine learning models on small datasets. Due to the high costs associated with labeling data, in Software Engineering, there exist many small (< 5,000 samples) and...

Ausführliche Beschreibung

Gespeichert in:
Bibliographische Detailangaben
Veröffentlicht in:IEEE transactions on software engineering 2022-12, Vol.48 (12), p.5050-5067
Hauptverfasser: Prenner, Julian Aron, Robbes, Romain
Format: Artikel
Sprache:eng
Schlagworte:
active learning
back translation
BERT
Codes
data augmentation
Datasets
fine-tuning
Machine learning
pre-training
RoBERTA
Small datasets
soft labels
Software
Software engineering
Support vector machines
Task analysis
Training
transformer
Transformers
Online-Zugang:Volltext bestellen
Tags: Tag hinzufügen
Keine Tags, Fügen Sie den ersten Tag hinzu!
Beschreibung
Zusammenfassung:This paper provides a starting point for Software Engineering (SE) researchers and practitioners faced with the problem of training machine learning models on small datasets. Due to the high costs associated with labeling data, in Software Engineering, there exist many small (< 5,000 samples) and medium-sized (
ISSN:0098-5589
1939-3520
DOI:10.1109/TSE.2021.3135465