Software effort estimation as a multiobjective learning problem

Minku, Leandro L.; Yao, Xin

doi:10.1145/2522920.2522928

Cited by 82 publications

(82 citation statements)

References 53 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…The base learner considered is MLP. The results show clearly the advantages of the multi-objective approach over the existing work [7].…”

Section: Multi-objective Ensemble Learning For Software Effort Estimamentioning

confidence: 78%

“…Software effort estimation (SEE) can be formulated as a multi-objective learning problem, where different objectives correspond to different performance measures [7]. A multi-objective evolutionary algorithm (MOEA) is used to better understand the trade-off among different performance measures by creating SEE models through simultaneous optimisation of these measures.…”

Section: Multi-objective Ensemble Learning For Software Effort Estimamentioning

confidence: 99%

See 1 more Smart Citation

Some Recent Work on Multi-objective Approaches to Search-Based Software Engineering

Yao

2013

Search Based Software Engineering

Self Cite

View full text Add to dashboard Cite

Abstract. Multi-objective algorithms have been used to solve difficult software engineering problems for a long time. This article summarises some selected recent work of applying latest meta-heuristic optimisation algorithms and machine learning algorithms to software engineering problems, including software module clustering, testing resource allocation in modular software system, protocol tuning, Java container testing, software project scheduling, software project effort estimation, and software defect prediction. References will be given, from which the details of such application of computational intelligence techniques to software engineering problems can be found.

show abstract

“…The base learner considered is MLP. The results show clearly the advantages of the multi-objective approach over the existing work [7].…”

Section: Multi-objective Ensemble Learning For Software Effort Estimamentioning

confidence: 78%

Section: Multi-objective Ensemble Learning For Software Effort Estimamentioning

confidence: 99%

Some Recent Work on Multi-objective Approaches to Search-Based Software Engineering

Yao

2013

Search Based Software Engineering

Self Cite

View full text Add to dashboard Cite

show abstract

“…While most of existing work in this space use single-objective search, a few of them (e.g. [26,28]) has recently proposed multi-objective search approach to e↵ort estimation. For example, a recent study done by Sarro et.…”

Section: Related Workmentioning

confidence: 99%

Multi-objective search-based approach to estimate issue resolution time

Al-Zubaidi

Dam

Ghose

et al. 2017

Proceedings of the 13th International Conference on Predictive Models and Data Analytics in Software Engineering

View full text Add to dashboard Cite

Background: Resolving issues is central to modern agile software development where a software is developed and evolved incrementally through series of issue resolutions. An issue could represent a requirement for a new functionality, a report of a software bug or a description of a project task. Aims: Knowing how long an issue will be resolved is thus important to different stakeholders including end-users, bug reporters, bug triagers, developers and managers. This paper aims to propose a multi-objective search-based approach to estimate the time required for resolving an issue. Methods: Using genetic programming (a meta-heuristic optimization method), we iteratively generate candidate estimate models and search for the optimal model in estimating issue resolution time. The search is guided simultaneously by two objectives: maximizing the accuracy of the estimation model while minimizing its complexity. Results: Our evaluation on 8,260 issues from five large open source projects demonstrate that our approach significantly (p < 0.001) outperforms both the baselines and state-of-the-art techniques. Conclusions: Evolutionary search-based approaches offer an effective alternative to build estimation models for issue resolution time. Using multiple objectives, one for measuring the accuracy and the other for the complexity, helps produce accurate and simple estimation models. Resolving issues is central to modern agile software development where a software is developed and evolved incrementally through series of issue resolutions. An issue could represent a requirement for a new functionality, a report of a software bug or a description of a project task.Aims: Knowing how long an issue will be resolved is thus important to di↵erent stakeholders including end-users, bug reporters, bug triagers, developers and managers. This paper aims to propose a multi-objective search-based approach to estimate the time required for resolving an issue.Methods: Using genetic programming (a meta-heuristic optimization method), we iteratively generate candidate estimate models and search for the optimal model in estimating issue resolution time. The search is guided simultaneously by two objectives: maximizing the accuracy of the estimation model while minimizing its complexity.Results: Our evaluation on 8,260 issues from five large open source projects demonstrate that our approach significantly (p < 0.001) outperforms both the baselines and state-of-the-art techniques.Conclusions: Evolutionary search-based approaches o↵er an e↵ective alternative to build estimation models for issue resolution time. Using multiple objectives, one for measuring the accuracy and the other for the complexity, helps produce accurate and simple estimation models.

show abstract

“…Much of the previous work on search-based project management (Aguilar-Ruiz et al 2001) (Alba and Chicano 2005) (Alba and Chicano 2007) (Chang 1994) ) (Chang et al 1998) (Chang et al 2001) (Chao et al 1993) (Minku et al 2012) ) has used synthetic data. This can be achieved in a disciplined and controlled manner.…”

Section: Minimising Software Project Completion Timementioning

confidence: 99%

Search-Based Software Project Management

Ferrucci

Harman

Sarro

2014

Software Project Management in a Changing World

View full text Add to dashboard Cite

Abstract. Project management presents the manager with a complex set of related optimisation problems. Decisions made can more profoundly affect the outcome of a project than any other activity. In the chapter, we provide an overview of Search-Based Software Project Management, in which Search-Based Software Engineering (SBSE) is applied to problems in software project management. We show how SBSE has been used to attack the problems of staffing, scheduling, risk, and effort estimation. SBSE can help to solve the optimisation problems the manager faces, but it can also yield insight. SBSE therefore provides both decision making and decision support. We provide a comprehensive survey of Search-Based Software Project Management, and give directions for the development of this subfield of SBSE. IntroductionSoftware Project Management includes several activities critical for the success of a project (e.g., cost estimation, project planning, quality management). These activities often involve finding a suitable balance between competing and potentially conflicting goals. For example, planning a project schedule requires to minimise the project duration and the project cost, and to maximise the product quality. Many of these problems are essentially optimization questions characterised by competing goals/constraints and with a bewilderingly large set of possible choices. So finding good solutions can be hard.Search-Based Software Engineering seeks to reformulate software engineering problems as search-based optimisation problems and applies a variety of metaheuristics based on local and global search to solve them (such as Hill Climbing, Tabu Search, and Genetic Algorithms). These meta-heuristics search for a suitable solution in a typically large input space guided by a fitness function that expresses the goals and leads the exploration into potentially promising areas of the search space.Though the term Search-Based Software Engineering (SBSE) was coined by Harman and Jones in 2001 to cover the application of computational search and optimisation across the wide spectrum of software engineering activities (Harman and Jones 2001), there were already pockets of activity on several specific software engineering problems prior to the introduction of the term SBSE. One such topic was Search-Based Software Project Management, the topic of this chapter. In particular, there was work on search-based project scheduling and staffing by

show abstract

Software effort estimation as a multiobjective learning problem

Cited by 82 publications

References 53 publications

Some Recent Work on Multi-objective Approaches to Search-Based Software Engineering

Some Recent Work on Multi-objective Approaches to Search-Based Software Engineering

Multi-objective search-based approach to estimate issue resolution time

Search-Based Software Project Management

Contact Info

Product

Resources

About