Jove
Visualize
Contact Us
JoVE
x logofacebook logolinkedin logoyoutube logo
ABOUT JoVE
OverviewLeadershipBlogJoVE Help Center
AUTHORS
Publishing ProcessEditorial BoardScope & PoliciesPeer ReviewFAQSubmit
LIBRARIANS
TestimonialsSubscriptionsAccessResourcesLibrary Advisory BoardFAQ
RESEARCH
JoVE JournalMethods CollectionsJoVE Encyclopedia of ExperimentsArchive
EDUCATION
JoVE CoreJoVE BusinessJoVE Science EducationJoVE Lab ManualFaculty Resource CenterFaculty Site
Terms & Conditions of Use
Privacy Policy
Policies

Related Experiment Video

Updated: Jan 12, 2026

Design and Analysis for Fall Detection System Simplification
08:05

Design and Analysis for Fall Detection System Simplification

Published on: April 6, 2020

11.1K

Deep multi-metrics learning for mobile app defect prediction using code and process metrics.

Ahmed Abdu1, Hakim A Abdo2, Inam Ullah3

  • 1School of Information, Xi'an University of Finance and Economics, Xi'an, 710100, China.

Scientific Reports
|November 4, 2025
PubMed
Summary

Related Concept Videos

You might also read

Related Articles

Articles linked to this work by shared authors, journal, and citation graph.

Sort by
Same author

Smart skies: optimizing autonomous AUAV positioning for robust IoT connectivity in next-gen cities.

Scientific reports·2026
Same author

Towards employee-centric HR: Understanding the link between opportunity-enhancing HR practices and employee taking charge at work.

Work (Reading, Mass.)·2026
Same author

Systematic characterization of purple acid phosphatases provides targets for improving phosphorus use efficiency in cotton.

Functional & integrative genomics·2026
Same author

Design and development of a 1-bit dual-mode metasurface for sub-6 GHz wireless communication systems.

Scientific reports·2026
Same author

AI-driven neuroanalytic modeling for mental health: multichannel CNN-based autism spectrum disorder detection via facial pattern analysis.

Frontiers in computational neuroscience·2026
Same author

An interpretable deep concatenated architecture for osteoporosis detection using enhanced knee radiographs.

Frontiers in medicine·2026

This study introduces a Deep Multi-Metrics Learning Model (DMLM) for accurate mobile app defect prediction. DMLM effectively combines code and process metrics, outperforming existing methods in both effort-aware and non-effort-aware scenarios.

Area of Science:

  • Software Engineering
  • Machine Learning
  • Mobile Application Development

Background:

  • Accurate defect prediction is crucial for efficient mobile app development and resource allocation.
  • Existing defect prediction models are limited by their reliance on isolated data sources, failing to capture the interplay of code and process metrics.
  • There is a need for advanced models that integrate diverse metrics for improved prediction accuracy.

Purpose of the Study:

  • To propose a novel Deep Multi-Metrics Learning Model (DMLM) for enhanced mobile app defect prediction.
  • To leverage both code metrics from the current version and process metrics from previous releases within a unified framework.
  • To evaluate the performance of DMLM against state-of-the-art approaches.

Main Methods:

Keywords:
Code metricsDeep neural networkMobile app defect predictionProcess metrics

More Related Videos

Quantified Assessment of Infant's Gross Motor Abilities Using a Multisensor Wearable
09:24

Quantified Assessment of Infant's Gross Motor Abilities Using a Multisensor Wearable

Published on: May 17, 2024

2.1K

Related Experiment Videos

Last Updated: Jan 12, 2026

Design and Analysis for Fall Detection System Simplification
08:05

Design and Analysis for Fall Detection System Simplification

Published on: April 6, 2020

11.1K
Quantified Assessment of Infant's Gross Motor Abilities Using a Multisensor Wearable
09:24

Quantified Assessment of Infant's Gross Motor Abilities Using a Multisensor Wearable

Published on: May 17, 2024

2.1K
  • Developed a Deep Multi-Metrics Learning Model (DMLM) integrating code and process metrics.
  • Utilized a deep convolutional neural network (CNN) to identify complex patterns within the combined metrics.
  • Conducted experiments on nine real-world Android applications from Git repositories.

    • Main Results:

    • DMLM significantly outperformed state-of-the-art approaches in non-effort-aware settings, showing superior Area Under the Curve (AUC), F1 scores, and Matthews correlation coefficient (MCC).
    • The DMLM model also demonstrated superior performance compared to baseline methods in effort-aware scenarios (PofB20).
    • The findings confirm the model's effectiveness in diverse prediction contexts.

    Conclusions:

    • The proposed DMLM effectively enhances mobile app defect prediction by integrating code and process metrics.
    • DMLM offers a robust solution for improving resource allocation and software quality in mobile app development.
    • The study underscores the importance of multi-metric learning for accurate and reliable defect prediction.