Where should the bugs be fixed? More accurate information retrieval-based bug localization based on bug reports

By Jian Zhou, Hongyu Zhang, David Lo. International Conference on Software Engineering, 2012

686 citations recorded by the source corpus.

For a large and evolving software system, the project team could receive a large number of bug reports. Locating the source code files that need to be changed in order to fix the bugs is a challenging task. Once a bug report is received, it is desirable to automatically point out to the files that developers should change in order to fix the bug. In this paper, we propose BugLocator, an information retrieval based method for locating the relevant files for fixing a bug. BugLocator ranks all files based on the textual similarity between the initial bug report and the source code using a revised Vector Space Model (rVSM), taking into consideration information about similar bugs that have been fixed before. We perform large-scale experiments on four open source projects to localize more than 3,000 bugs. The results show that BugLocator can effectively locate the files where the bugs should be fixed. For example, relevant buggy files for 62.60% Eclipse 3.1 bugs are ranked in the top ten among 12,863 files. Our experiments also show that BugLocator outperforms existing state-of-the-art bug localization methods.

Subfields: Information Systems, Software

View Paper: https://www.semanticscholar.org/paper/3f3816ca55d4681739f9dd920fdb7fe3d190b0b3

How researchers see this paper

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

In Information Systems

In Software

Citations in the corpus

From Word Embeddings to Document Similarities for Improved Information Retrieval in Software Engineering (2016, methodology): “[47] and the LDA-based BugScout from Nguyen et al.”
Agentless: Demystifying LLM-based Software Engineering Agents (2024, methodology): “A GENTLESS ’s localization process make uses of both LLM-based localization as well classic information-retrieval-based localization idea [109].”
Practitioners' expectations on automated fault localization (2016, methodology): “They analyzed what information in a bug report tends to produce good results, how their user study participants used information in bug reports, and whether the participants behaved diﬀerently when they use BugLocator than without it.”
An Empirical Study on Learning Bug-Fixing Patches in the Wild via Neural Machine Translation (2018, methodology): “osed of five collections of source code files taken from real commits in open-source projects, belonging to seven different studies in the literature focusing on bug-fixes and change history analysis [28, 46, 55, 71, 79, 95, 97]. On theCodRep dataset we perform the same steps described in Section2.2”
A Survey on Software Fault Localization (2016, result): “Recently, some studies [85], [234], [298], [315], [425] have applied information retrieval techniques to software fault localization.”
A Novel Neural Source Code Representation Based on Abstract Syntax Tree (2019, background): “represented by token sequences or bag of tokens for code clone detection [3], [4], bug localization[11], and code authorship”
Easy over hard: a case study on deep learning (2017, result): “By comparing it to baseline methods (Naive Bayes [32], learn-to-rank [76], BugLocator [79]), Lam et al. reported, 16.2-46.4%, 8-20.8% and 2.7-20.7% higher top-1 accuracy than baseline methods, respectively [39].”
Learning to rank relevant files for bug reports using domain knowledge (2014, methodology): “Recently, researchers [19, 25, 34, 40, 41, 44, 45] have developed methods that concentrate on ranking source (cid:12)les for given bug reports automatically.”
Improving bug localization using structured information retrieval (2013, methodology): “[46], enabling fair comparison and reproducible findings.”
Scaffle: bug localization on millions of files (2020, methodology): “For example, as shown in Figure 6(a), the top-5 predictions have a higher hit rate than the heuristic baseline.”
SHERLOC: Structured Diagnostic Localization for Code Repair Agents (2026, unknown): “Most localization methods retrieve faulty files and functions [Zhou et al., 2012, Reddy et al., 2025, Yu et al., 2025]; some (e.g., OrcaLoca, SWE-Debate [Li et al., 2025]) also produce free-form report text.”
Agentic Software Issue Resolution with Large Language Models: A Survey (2025, unknown): “Embedding-based methods [164, 196] identify suspicious code snippets by projecting both issues and code representations into a shared semantic space, thereby enabling similarity-based matching for accurate localization.”