On the Accuracy of Spectrum-based Fault Localization

By Rui Abreu, Peter Zoeteweij, A. V. Gemund. Testing: Academic and Industrial Conference Practice and Research Techniques - MUTATION (TAICPART-MUTATION 2007), 2007

898 citations recorded by the source corpus.

Subfields: Computer Networks and Communications, Hardware and Architecture, Software

View Paper: https://www.semanticscholar.org/paper/eb956ab442810f783f44600ce6d714583a96a4aa

How researchers see this paper

The researchers whose trusted networks rank it highest, strongest advocate first. Every researcher gets a different order.

In Computer Networks and Communications

In Hardware and Architecture

In Software

Citations in the corpus

A syntax-guided edit decoder for neural program repair (2021, methodology): “Recoder uses Ochiai [1] (implemented in GZoltar [57]), which is widely used in existing APR tools [24, 40].”
Precise Condition Synthesis for Program Repair (2016, methodology): “The SBFL formula we used in our implementation is Ochiai [28], which is shown to be among the most effective spectrum-based fault localization algorithms [29], [30], [31].”
A theoretical analysis of the risk evaluation formulas for spectrum-based fault localization (2013, background): “… Technologies, Swinburne University of Technology, John St Hawthorn VIC 3122, Australia; email: xxie@swin.edu.au; B. Xu, State Key Laboratory for Novel Software Technology and Department of Computer Science and Technology, Nanjing University, Huiwen Rd Gulou, Nanjing, Jiangsu, China.”
Scaffle: bug localization on millions of files (2020, methodology): “Overall, Scaffle provides a practical technique for pinpointing the files to consider in a large-scale code base, which helps assigning crashes to the appropriate team or developer, and may serve as a starting point for automated program repair.”
AutoCodeRover: Autonomous Program Improvement (2024, neutral): “If a generated patch does not follow the specified patch format or could not be applied syntactically to the original program, the agent is prompted to retry.”
Less training, more repairing please: revisiting automated program repair via zero-shot learning (2022, methodology): “We use Ochiai fault localization [3], same as previous approaches [44, 80].”
A Roadmap Toward the Resilient Internet of Things for Cyber-Physical Systems (2018, methodology): “a slicing method [203] that maps these time segments to the internal states and transitions of the model and a spectrumbased fault-localization method [139] that produces a ranking of the internal states and/or transitions that are most likely to explain the fault.”
Automated Program Repair in the Era of Large Pre-trained Language Models (2022, neutral): “While AlphaRepair is able to achieve state-of-the-art results, CodeBERT is considerably smaller than the newest LLMs. Additionally, AlphaRepair is designed for the repair setting where the buggy line location is known (e.g., computed by fault localization techniques [46]).”
Agentless: Demystifying LLM-based Software Engineering Agents (2024, methodology): “Dynamic FL techniques mainly include spectrum-based FL (SBFL) [47, 25, 24] and mutation-based FL (MBFL) [77, 68, 63].”
Evaluating and Improving Fault Localization (2017, methodology): “[18] C Tarantula Siemens 3 122 x [3] C Ochiai, Tarantula Siemens 3 120 x [2] C Ochiai, Tarantula Siemens, space 12 128 x 34 [4] C Barinel, Ochiai, Tarantula Siemens, space, gzip, sed 31 141 x 38 [5] C Tarantula Concordance 2 200 13 [31] C Op2, Ochiai, Tarantula Siemens, space 12 132 x 32 [33] C Meta”
Spectrum-Based Multiple Fault Localization (2009, methodology): “SFL uses abstraction of program traces to correlate software component activity with program failures (a statistical approach) [4], [14], [20], [24], [29], [37].”
Metallaxis‐FL: mutation‐based fault localization (2015, methodology): “Researchers have used many coverage elements such as program statements [9], [10], [11] program branches [12], [8] program du-pairs [13] and possible combinations of them [8], [14].”
A Survey on Software Fault Localization (2016, methodology): “Many studies that evaluate the effectiveness of speciﬁc fault localization techniques have been reported [8], [10], [11], [12], [25], [31], [36], [49], [53], [91], [93], [94], [102], [124], [178], [185], [191], [207], [209], [253], [266], [267], [296], [299], [335], [366], [390], [391], [393],…”
A practical evaluation of spectrum-based fault localization (2009, result): “Preliminary results of Section 4 appeared in (Abreu et al., 2006b), and of Sections 5 and 6 in (Abreu et al., 2007).”
CODIT: Code Editing With Tree-Based Neural Models (2018, methodology): “Here we assume that a fault-localization technique already localizes the bug [51].”
A model for spectra-based software diagnosis (2011, methodology): “In previous work [Abreu et al. 2007], there is empirical evidence that around .ve failed test cases is suf.cient for good performance of Ochiai, and increasing the number of passed tests beyond 20 results in only minor improvement in performance.”
History Driven Program Repair (2016, methodology): “We presently assume that the faulty methods are known in advance, as file- and method-level localization represent an orthogonal problem; we then compute the faulty lines in each prospective faulty method using existing statistical fault localization techniques [2].”