Modelling the ‘hurried’ bug report reading process to summarize bug reports

Lotufo, Rafael; Malik, Zeeshan Haider; Czarnecki, Krzysztof

doi:10.1007/s10664-014-9311-2

Cited by 71 publications

(74 citation statements)

References 32 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Mani et al [19] have used centroid, Maximum Marginal Relevance, Grasshopper and DiverseRank unsupervised techniques for summarizing bug reports. Lotufo et al [4] used unsupervised bug reports summarization to generate the summaries and used PageRank to calculate the probabilities. They used number of topics shared between the sentences, number of times sentence evaluated by other sentences and number of topics shared with title and description of bug report to ranks the sentences from relevance perspective.…”

Section: Extractive Summarization: An Overview In the Context Of Softmentioning

confidence: 99%

“…They used 3-scale likert scale to rate the summaries in terms of conciseness, content adequacy and accuracy. Lotufo et al [4] in the first part of their evaluation used intrinsic evaluations and compared summaries produced by Rastkar et al [13] by using precision, recall, pyramid score and F-score.…”

Section: Pyramid Score=a \ Bmentioning

confidence: 99%

“…Bug reports not just contain the defects but also the reasons behind the bugs, feature enhancement ideas, steps for resolution of bugs, etc. [4]. Bug reports summarization helps user read, investigate and understand multiple aspects of a defect and helps perform many tasks like bug reports duplication, etc.Low time to market and tough competition have raised the need of summarization tools to help developers.…”

Section: Introductionmentioning

confidence: 99%

See 2 more Smart Citations

Summarization of Software Artifacts : A Review

Gupta¹,

S.K²

2017

IJCSIT

View full text Add to dashboard Cite

Summarization of software artifacts is an ongoing field of research among the software engineering community due to the benefits that summarization provides like saving of time and efforts in various software engineering tasks like code search, duplicate bug reports detection, traceability link recovery, etc. Summarization is to produce short and concise summaries. The paper presents the review of the state of the art of summarization techniques in software engineering context. The paper gives a brief overview to the software artifacts which are mostly used for summarization or have benefits from summarization. The paper briefly describes the general process of summarization. The paper reviews the papers published from 2010 to June 2017 and classifies the works into extractive and abstractive summarization. The paper also reviews the evaluation techniques used for summarizing software artifacts. The paper discusses the open problems and challenges in this field of research. The paper also discusses the future scopes in this area for new researchers.

show abstract

Section: Extractive Summarization: An Overview In the Context Of Softmentioning

confidence: 99%

Section: Pyramid Score=a \ Bmentioning

confidence: 99%

Section: Introductionmentioning

confidence: 99%

See 1 more Smart Citation

Summarization of Software Artifacts : A Review

Gupta¹,

S.K²

2017

IJCSIT

View full text Add to dashboard Cite

show abstract

“…For example, Latent Semantic Analysis (LSI) [Dum04] and vector space models [MRS08] are used to recover traceability links between documentation and source code [ACCL00,MM03], and several algorithms for textual summarization have been used to summarize development artifacts like bug reports [MCSD12,LMC12,RMM14a].…”

Section: Pluralitas Non Est Ponenda Sine Necessitate (Plurality Shoulmentioning

confidence: 99%

Holistic recommender systems for software engineering

Ponzanelli

2014

Companion Proceedings of the 36th International Conference on Software Engineering

View full text Add to dashboard Cite

“…Finally, several researchers have investigated text to text summarization, the summarization of software textual artifacts such as bug reports [50,54,74]. The earliest work was the application of single-document extractive summarization on bug reports by Rastkar et al [74].…”

Section: Summarization Of Software Artifactsmentioning

confidence: 99%

Code fragment summarization

Ying

Robillard

2013

Proceedings of the 2013 9th Joint Meeting on Foundations of Software Engineering

View full text Add to dashboard Cite

Code fragments are an important resource for understanding the Application Programming Interface (API) of software libraries. Many usage scenarios for code fragments require them to be distilled to their essence: for example, when serving as cues to longer documents, for reminding developers of a previously known idiom, or for displaying search results. This dissertation reports on research on shortening, or summarizing, code fragments and makes three main contributions: a set of lessons learned from a case study on a supervised machine learning approach to the generation of code fragment summaries; an empirically grounded catalog of source code summarization practices; and the design, implementation and evaluation of a novel optimization-based summarization technique for code fragments.The case study on the generation of code fragment summaries was based on a supervised machine learning approach that classifies whether a line in a code fragment should be in a summary. We present the lessons learned that were key to the two subsequent parts of the research: the best performing feature set being a combination of syntactic and query-related features, and three limitations on our supervised machine learning approach and the line-based problem formulation. The limitations were in using line as the granularity, obtaining training data with high quality, and using only features that are local to a line without considering dependencies among different parts of the code.Motivated by the limitations of line-based summaries, we studied how humans shorten code fragments to understand the nature of the output of the summarization process. Based on 156 hand-generated summaries obtained i from 16 participants, we analyzed decisions on which content to select and how to present this content in a summary. We elicited a catalog of common summarization practices behind these decisions across the summaries, as well as the rationale behind the practices, using a mix of qualitative and quantitative methods. We found that none of the participants exclusively extracted code verbatim for the summaries. Participants employed many practices to modify the content, by trimming a line, truncating code, aggregating a large amount of code, and refactoring code. Not only were the participants concerned with the main goal of the task to shorten code, but also with whether the summary looked compilable, readable and understandable.With the insights from the machine learning case study and the catalog of summarization practices, we devised a technique to generate summaries constrained in both height and width: given as input a code fragment and a query (a set of keywords), our technique produces a shorter version of the fragment that fits in a two-dimensional space (L lines by W columns) and that captures as much as possible of the essential elements of the original code related to the query, while remaining readable. To generate these summaries, we developed a code summarization tool called Konaila. Konaila maximizes the value of the con...

show abstract

Modelling the ‘hurried’ bug report reading process to summarize bug reports

Cited by 71 publications

References 32 publications

Summarization of Software Artifacts : A Review

Summarization of Software Artifacts : A Review

Holistic recommender systems for software engineering

Code fragment summarization

Contact Info

Product

Resources

About