Software Engineering Meets Deep Learning: A Mapping Study

Ferreira, Fabio; Silva, Luciana Lourdes; Valente, Marco Tulio

doi:10.48550/arxiv.1909.11436

Cited by 3 publications

(4 citation statements)

References 83 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…They also briefly discuss an overview of concerns when using DL within SE, largely drawn from existing work. Similarly, Ferreira et al [46] provide similar analysis to the paper presented by Li et al However, Ferreira et al provides a brief description of the works they studied as well as highlights some of their strengths in regards to the specific tasks they addressed. They also perform a general survey of the type of DL architectures implemented.…”

Section: Related Studies and Literature Reviewsmentioning

confidence: 84%

“…The most highly related works to the work presented in this study also looks at the application of DL to a variety of SE tasks. The two most closely related papers in this space are non-peer reviewed literature reviews hosted on arXiv by Li et al [82] and Ferrerire et al [46], which we briefly discuss here for completeness. Li et al's study analyzes 98 DL studies for general trends and applications in DL.…”

Section: Related Studies and Literature Reviewsmentioning

confidence: 99%

See 1 more Smart Citation

A Systematic Literature Review on the Use of Deep Learning in Software Engineering Research

Watson¹,

Cooper²,

Palacio³

et al. 2020

Preprint

View full text Add to dashboard Cite

An increasingly popular set of techniques adopted by software engineering (SE) researchers to automate development tasks are those rooted in the concept of Deep Learning (DL). The popularity of such techniques largely stems from their automated feature engineering capabilities, which aid in modeling software artifacts. However, due to the rapid pace at which DL techniques have been adopted it is difficult to distill the current successes, failures, and opportunities of the current research landscape. In an effort to bring clarity to this crosscutting area of work, from its modern inception to the present, this paper presents a systematic literature review of research at the intersection of SE & DL. The review canvases work appearing in the most prominent SE and DL conferences and journals and spans 84 papers across 22 unique SE tasks. We center our analysis around the components of learning, a set of principles that govern the application of machine learning techniques (ML) to a given problem domain, discussing several aspects of the surveyed work at a granular level. The end result of our analysis is a research roadmap that both delineates the foundations of DL techniques applied to SE research, and likely areas of fertile exploration for the future.CCS Concepts: • Software and its engineering → Software creation and management; Software development techniques;

show abstract

Section: Related Studies and Literature Reviewsmentioning

confidence: 84%

Section: Related Studies and Literature Reviewsmentioning

confidence: 99%

A Systematic Literature Review on the Use of Deep Learning in Software Engineering Research

Watson¹,

Cooper²,

Palacio³

et al. 2020

Preprint

View full text Add to dashboard Cite

show abstract

“…To identify the relevant primary studies, I defined the inclusion and exclusion criteria listed in Table 4. To present a SE perspective and keep the size of the final primary study pool manageable, I only considered papers published in the SE journals and conference/workshop/symposium proceedings, as others have done [7].…”

Section: Goal and Review Questionsmentioning

confidence: 99%

A software engineering perspective on engineering machine learning systems: State of the art and challenges

Giray¹

2021

Journal of Systems and Software

103

View full text Add to dashboard Cite

“…Deep learning has been used in program repair using neural machine translation [5,30], sequence-editing approaches [23], and learning graph transformations [10]. For a deeper review of deep learning methods applied to software engineering tasks, see the literature reviews [11,21].…”

Section: Related Workmentioning

confidence: 99%

DeepMerge: Learning to Merge Programs

Dinella¹,

Mytkowicz²,

Svyatkovskiy³

et al. 2021

Preprint

View full text Add to dashboard Cite

Program merging is ubiquitous in modern software development.Although commonly used in most version control systems, textbased merge algorithms are prone to producing spurious merge conflicts: they report a conflict even when program changes do not interfere with each other semantically. Spurious merge conflicts are costly to development as the need for manual intervention stalls modern continuous integration pipelines. We propose a novel data-driven approach to identify and resolve spurious merge conflicts with a sequence-to-sequence machine learning model. We realize our approach in a tool DeepMerge that uses a novel combination of (i) an edit-aware embedding of merge inputs and (ii) a variation of pointer networks to construct resolutions from input segments. We also propose an algorithm to extract ground truth manual resolutions from a code corpus and employ it to curate a dataset comprising 10,729 non-trivial resolutions in Javascript programs. Our evaluation shows that DeepMerge can predict correct resolutions with high precision (72%) and modest recall (34%) on the dataset overall, and high recall (78%) on merges comprising of upto 3 lines that comprise 24% of the dataset.

show abstract

Software Engineering Meets Deep Learning: A Mapping Study

Cited by 3 publications

References 83 publications

A Systematic Literature Review on the Use of Deep Learning in Software Engineering Research

A Systematic Literature Review on the Use of Deep Learning in Software Engineering Research

A software engineering perspective on engineering machine learning systems: State of the art and challenges

DeepMerge: Learning to Merge Programs

Contact Info

Product

Resources

About