Machine Learning in Requirements Engineering: A Mapping Study

Zamani, Kareshna; Zowghi, Didar; Arora, Chetan

doi:10.1109/rew53955.2021.00023

Cited by 17 publications

(7 citation statements)

References 22 publications

Supporting

Mentioning

Contrasting

Unclassified

Order By: Relevance

“…Paradoxically, although machine learning technique is used extensively in the actual development of software; it's minimal application in requirements engineering tools was an interesting observation. [36] noted that the major challenge of using ML approach in requirements engineering was the lack of relevant datasets. This contributing factor is similar to the analysis of RQ3.…”

Section: Discussionmentioning

confidence: 99%

A Review and Analysis of Tools Used from 2018 till 2022 in Requirements Engineering

Yumun¹,

Embi²

2022

Proceedings of the International Conference on Computer, Information Technology and Intelligent Computing (CITIC 2022)

View full text Add to dashboard Cite

The commonly known stages in Requirements Engineering (RE) include planning, elicitation, documentation, validation, and management. Each stage consists of multiple processes and activities between stakeholders that result in data collection and analysis. Activities and data can be collected/extracted/analysed by manual or automated software tools to streamline the requirements engineering process. The objective of this review is to collate and study the distribution of tools in each stage and to discover missing gaps (if any) in tools for each stage. An electronic search was conducted on four publication libraries to compile all tools introduced in the requirements engineering field from 2018 to April 2022. 33 publications were found, and data extraction was performed on them to classify the tools to discover common patterns and methodologies. The results from data analysis showed that most new tools focused on elicitation and validation stages of RE. Planning and documentation stages had the least number of new tools. No new methodologies for RE tool creation were introduced. NLP remained the popular approaches among researchers. 21 out of 33 publications did not present case studies to prove the practical applications of the tools they introduced. This review has aggregated knowledge in the tools used in different RE stages.

show abstract

Section: Discussionmentioning

confidence: 99%

A Review and Analysis of Tools Used from 2018 till 2022 in Requirements Engineering

Yumun¹,

Embi²

2022

Proceedings of the International Conference on Computer, Information Technology and Intelligent Computing (CITIC 2022)

View full text Add to dashboard Cite

show abstract

“…Here ML is used to support a range of tasks including: volatility prediction [10]; reuse of requirements and feature and variability extraction in the context of software product lines [20]; requirements elicitation, analysis, specification, prioritization, and negotiation [2,8,45]; and requirements ambiguity resolution [177].…”

Section: Software Requirementsmentioning

confidence: 99%

“…The following research gaps are observed in this area. Further empirical research is needed on how to assess and select the most suitable ML techniques in requirements volatility prediction and ambiguity resolution [10,177], and how to automate with ML the extraction of software requirements from natural language documents [8,20]. Researchers should experiment with ML in more requirements activities, such as requirements specifications and management [8].…”

Section: Software Requirementsmentioning

confidence: 99%

“…Researchers should experiment with ML in more requirements activities, such as requirements specifications and management [8]. Some research ideas include sharing standard pre-labeled datasets, standardizing nonfunctional requirements, applying sentiment analysis on functional and nonfunctional requirements, and performing change impact analysis in ML-based software requirements [2,177].…”

Section: Software Requirementsmentioning

confidence: 99%

See 1 more Smart Citation

Machine Learning for Software Engineering: A Tertiary Study

2023

View full text Add to dashboard Cite

Machine learning (ML) techniques increase the effectiveness of software engineering (SE) lifecycle activities. We systematically collected, quality-assessed, summarized, and categorized 83 reviews in ML for SE published between 2009–2022, covering 6 117 primary studies. The SE areas most tackled with ML are software quality and testing, while human-centered areas appear more challenging for ML. We propose a number of ML for SE research challenges and actions including: conducting further empirical validation and industrial studies on ML; reconsidering deficient SE methods; documenting and automating data collection and pipeline processes; reexamining how industrial practitioners distribute their proprietary data; and implementing incremental ML approaches.

show abstract

“…Recent studies have shown that many AI-based systems lack requirements specifications [73,74,75], which is mainly due to the difference in the building process between traditional systems vs. AI-based software [76]. We conducted an SLR [28] to identify literature that focused on using RE4AI.…”

Section: Related Workmentioning

confidence: 99%

Requirements Framework for Engineering Human-centered Artificial Intelligence-Based Software Systems

Ahmad¹,

Abdelrazek²,

Arora³

et al. 2023

Preprint

View full text Add to dashboard Cite

Artificial intelligence (AI) components used in building software solutions have substantially increased in recent years. However, many of these solutions end up focusing on technical aspects and ignore critical human-centered aspects. [Objective] Including humancentered aspects during requirements engineering (RE) when building AI-based software can help achieve more responsible, unbiased, and inclusive AI-based software solutions. [Method] In this paper, we present a new framework developed based on human-centered AI guidelines and a user survey to aid in collecting requirements for human-centered AI-based software. We provide a catalog to elicit these requirements and a conceptual model to present them visually.[Results] The framework is applied to a case study to elicit and model requirements for enhancing the quality of 360°videos intended for virtual reality (VR) users. [Conclusion] We found that our proposed approach helped the project team fully understand the needs of the project to deliver. Furthermore, the framework helped to understand what requirements need to be captured at the initial stages against later stages in the engineering process of AI-based software.

show abstract

Machine Learning in Requirements Engineering: A Mapping Study

Cited by 17 publications

References 22 publications

A Review and Analysis of Tools Used from 2018 till 2022 in Requirements Engineering

A Review and Analysis of Tools Used from 2018 till 2022 in Requirements Engineering

Machine Learning for Software Engineering: A Tertiary Study

Requirements Framework for Engineering Human-centered Artificial Intelligence-Based Software Systems

Contact Info

Product

Resources

About