A Survey of Challenges in Spectrum-Based Software Fault Localization

Sarhan, Qusay Idrees; Beszedes, Arpad

doi:10.1109/access.2022.3144079

Cited by 34 publications

(10 citation statements)

References 153 publications

Supporting

Mentioning

Contrasting

Unclassified

Order By: Relevance

“…In recent years, researchers have introduced numerous fault localization methods to pinpoint the root causes of failures in traditional software programs. These methods fall into eight main categories [18]: spectrum-based [19], [20], slice-based [21], [22], statistics-based [23], [24], program state-based [25], machine learning-based [26], [27], modelbased [28], data mining-based [29], and miscellaneous techniques [30]. Numerous studies have been undertaken to assess the efficiency of these fault localization techniques.…”

Section: Related Workmentioning

confidence: 99%

DeepCNN: A Dual Approach to Fault Localization and Repair in Convolutional Neural Networks

Wardat,

Al-Alaj

2024

IEEE Access

View full text Add to dashboard Cite

Deep learning models, particularly Convolutional Neural Networks (CNNs), play a pivotal role in intelligent software. However, like any software application, CNN-based applications are susceptible to bugs. Bug-fix patterns in CNN differ from traditional techniques, primarily due to their inherent blackbox nature. Moreover, current methods, although tailored for generic DNN structures, are time consuming, require specialized expertise, and are not directly applicable to the unique structure and requirements of CNNs. To address these issues, we propose DeepCNN, an innovative automated tool to identify the root causes of CNN faults and autonomously fix prevalent training faults that impact the performance and efficiency of CNN programs. DeepCNN, a data-driven tool, employs a transformer encoder model to abstract token-level CNN code, enabling it to effectively detect and fix bugs in real-world CNN models. Beyond mere identification, DeepCNN offers automated solutions for repairing CNN hyperparameter misconfigurations using an optimization solver. Additionally, by analyzing data dependencies and employing a search algorithm, it determines the most optimal hyperparameter values for the problematic models, ultimately generating patch fixes. Through rigorous evaluation on 36 diverse buggy models, DeepCNN outperformed existing methods in fault localization and repair, showing a robustness in identifying potential problems with a 90% detection rate. Among these problematic models, it manages to repair 90% of them -resulting in a 30% average accuracy boost.

show abstract

Section: Related Workmentioning

confidence: 99%

DeepCNN: A Dual Approach to Fault Localization and Repair in Convolutional Neural Networks

Wardat,

Al-Alaj

2024

IEEE Access

View full text Add to dashboard Cite

show abstract

“…In this chapter, the following points summarize my main contributions to the topic of thesis point I. The results of this chapter were published in [122].…”

Section: Contributionsmentioning

confidence: 99%

“…Ebben a fejezetben a következő pontok foglalják össze az I. tézispont témájával kapcsolatos főbb eredményeimet [122].…”

Section: Abbreviationsunclassified

“…To help developers discover the faulty element early in the examination process and with minimal effort, the faulty element should be put near the highest place in the ranking list. However, ranking based only on suspiciousness scores inevitably involves a problem called rank ties [122]. When different code elements are tied this means that they have the same suspiciousness scores, so they are indistinguishable from each other in this respect.…”

Section: Introductionmentioning

confidence: 99%

“…This score can be used in debugging to identify the faulty element more efficiently. One of the main challenges in SBFL is how to introduce new formulas to enhance SBFL's performance by putting faulty elements at the beginning of the ranking list produced by SBFL as much as possible [122]. Thus, they can be examined and found efficiently.…”

Section: Introductionmentioning

confidence: 99%

See 2 more Smart Citations

Methods for Enhancing Software Fault Localization

Alsarhan

View full text Add to dashboard Cite

First of all, I would like to praise and thank God for everything. Then, I would like to thank my supervisor, Dr. Árpád Beszédes, for guiding me and directing my PhD study. I will always be thankful to him and will never forget the great moments we have spent together during our journey. Also, I would like to thank my colleagues and co-authors for their endless support. Of course, I would like to thank Edit Szűcs for correcting this thesis from a linguistic point of view. Last, but not least, I would like to thank my wife, children, parents, sister, and brothers for their constant love and support throughout my study. Without their support, my PhD study would not have been possible.Finally, I would like to thank the Department of Software Engineering of the University of Szeged for providing me with a productive study environment. The Stipendium Hungaricum Scholarship program funded my PhD study, so I am very appreciative of their support.

show abstract

Effective Spectrum Based Fault Localization Using Contextual Based Importance Weight

Sarhan

Beszedes

2022

Communications in Computer and Information Science

View full text Add to dashboard Cite

A Survey of Challenges in Spectrum-Based Software Fault Localization

Cited by 34 publications

References 153 publications

DeepCNN: A Dual Approach to Fault Localization and Repair in Convolutional Neural Networks

DeepCNN: A Dual Approach to Fault Localization and Repair in Convolutional Neural Networks

Methods for Enhancing Software Fault Localization

Effective Spectrum Based Fault Localization Using Contextual Based Importance Weight

Contact Info

Product

Resources

About