Enhancing the Software Defect Prediction using AutoInt

Shivani Jadhav; Saranya Manikandan; Duksan Ryu

doi:10.1109/BigComp64353.2025.00029

Enhancing the Software Defect Prediction using AutoInt

Shivani Jadhav^*
, Saranya Manikandan
, Duksan Ryu

^*Corresponding author for this work

Jeonbuk National University

Research output: Contribution to journal › Conference article › peer-review

2 Scopus citations

Abstract

Software defect prediction (SDP) is essential for ensuring software quality and reliability. Transformer architectures, known for their success in vision and NLP, offer significant potential for SDP. This study investigates the application of the AutoInt Transformer model, which excels in capturing complex feature interactions from numerical and categorical data, for predicting defects. AutoInt is evaluated on several SDP datasets and compared to advanced models like GRU, Tab-Net, FT-Transformer, and TabTransformer using standard evaluation metrics. Results demonstrate AutoInt's superior performance, further validated through effect size measurements. This research highlights the promise of transformer-based techniques for advancing SDP and enhancing software quality assurance.

Original language	English
Pages (from-to)	103-104
Number of pages	2
Journal	Proceedings of the IEEE International Conference on Big Data and Smart Computing, BIGCOMP
Issue number	2025
DOIs	https://doi.org/10.1109/BigComp64353.2025.00029
State	Published - 2025
Event	2025 IEEE International Conference on Big Data and Smart Computing, BigComp 2025 - Kota Kinabalu, Malaysia Duration: 2025.02.9 → 2025.02.12

Keywords

AutoInt
Deep Learning
Software Defect prediction
Transformer

Quacquarelli Symonds(QS) Subject Topics

Computer Science & Information Systems
Data Science

Access to Document

10.1109/BigComp64353.2025.00029

Cite this

@article{20156752900042d0be2cf34157a79cff,

title = "Enhancing the Software Defect Prediction using AutoInt",

abstract = "Software defect prediction (SDP) is essential for ensuring software quality and reliability. Transformer architectures, known for their success in vision and NLP, offer significant potential for SDP. This study investigates the application of the AutoInt Transformer model, which excels in capturing complex feature interactions from numerical and categorical data, for predicting defects. AutoInt is evaluated on several SDP datasets and compared to advanced models like GRU, Tab-Net, FT-Transformer, and TabTransformer using standard evaluation metrics. Results demonstrate AutoInt's superior performance, further validated through effect size measurements. This research highlights the promise of transformer-based techniques for advancing SDP and enhancing software quality assurance.",

keywords = "AutoInt, Deep Learning, Software Defect prediction, Transformer",

author = "Shivani Jadhav and Saranya Manikandan and Duksan Ryu",

note = "Publisher Copyright: {\textcopyright} 2025 IEEE.; 2025 IEEE International Conference on Big Data and Smart Computing, BigComp 2025 ; Conference date: 09-02-2025 Through 12-02-2025",

year = "2025",

doi = "10.1109/BigComp64353.2025.00029",

language = "English",

pages = "103--104",

journal = "Proceedings of the IEEE International Conference on Big Data and Smart Computing, BIGCOMP",

issn = "2375-933X",

number = "2025",

}

TY - JOUR

T1 - Enhancing the Software Defect Prediction using AutoInt

AU - Jadhav, Shivani

AU - Manikandan, Saranya

AU - Ryu, Duksan

PY - 2025

Y1 - 2025

N2 - Software defect prediction (SDP) is essential for ensuring software quality and reliability. Transformer architectures, known for their success in vision and NLP, offer significant potential for SDP. This study investigates the application of the AutoInt Transformer model, which excels in capturing complex feature interactions from numerical and categorical data, for predicting defects. AutoInt is evaluated on several SDP datasets and compared to advanced models like GRU, Tab-Net, FT-Transformer, and TabTransformer using standard evaluation metrics. Results demonstrate AutoInt's superior performance, further validated through effect size measurements. This research highlights the promise of transformer-based techniques for advancing SDP and enhancing software quality assurance.

AB - Software defect prediction (SDP) is essential for ensuring software quality and reliability. Transformer architectures, known for their success in vision and NLP, offer significant potential for SDP. This study investigates the application of the AutoInt Transformer model, which excels in capturing complex feature interactions from numerical and categorical data, for predicting defects. AutoInt is evaluated on several SDP datasets and compared to advanced models like GRU, Tab-Net, FT-Transformer, and TabTransformer using standard evaluation metrics. Results demonstrate AutoInt's superior performance, further validated through effect size measurements. This research highlights the promise of transformer-based techniques for advancing SDP and enhancing software quality assurance.

KW - AutoInt

KW - Deep Learning

KW - Software Defect prediction

KW - Transformer

UR - https://www.scopus.com/pages/publications/105006561086

U2 - 10.1109/BigComp64353.2025.00029

DO - 10.1109/BigComp64353.2025.00029

M3 - Conference article

AN - SCOPUS:105006561086

SN - 2375-933X

SP - 103

EP - 104

JO - Proceedings of the IEEE International Conference on Big Data and Smart Computing, BIGCOMP

JF - Proceedings of the IEEE International Conference on Big Data and Smart Computing, BIGCOMP

IS - 2025

T2 - 2025 IEEE International Conference on Big Data and Smart Computing, BigComp 2025

Y2 - 9 February 2025 through 12 February 2025

ER -

Enhancing the Software Defect Prediction using AutoInt

Abstract

Keywords

Quacquarelli Symonds(QS) Subject Topics

Access to Document

Other files and links

Fingerprint

Cite this