Software Effort Estimation Using Machine Learning Techniques

Sinha, Ripu Ranjan Sinha Ripu Ranjan; Gora, Rajani Kumari

doi:10.1007/978-981-15-5421-6_8

Cited by 9 publications

(5 citation statements)

References 15 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Humans have difficulty performing labelling because it is very costly and time-consuming. By assigning a pseudo-label to unlabelled data, semi-supervised algorithms generate output using unlabelled data [23]. Once the datasets have been labelled and pseudo-labelled, they can then be combined.…”

Section: Semi-supervised Learningmentioning

confidence: 99%

See 1 more Smart Citation

An Intelligent Framework for Estimating Software Development Projects using Machine Learning

Srivastava,

Agarwal

et al. 2023

IJRITCC

View full text Add to dashboard Cite

The IT industry has faced many challenges related to software effort and cost estimation. A cost assessment is conducted after software effort estimation, which benefits customers as well as developers. The purpose of this paper is to discuss various methods for the estimation of software effort and cost in the context of software engineering, such as algorithmic methods, expert judgment methods, analogy-based estimation methods, and machine learning methods, as well as their different aspects. In spite of this, estimation of the effort involved in software development are subject to uncertainty. Several methods have been developed in the literature for improving estimation accuracy, many of which involve the use of machine learning techniques. A machine learning framework is proposed in this paper to address this challenging problem. In addition to being completely independent of algorithmic models and estimation problems, this framework also features a modular architecture. It has high interpretability, learning capability, and robustness to imprecise and uncertain inputs.

show abstract

Section: Semi-supervised Learningmentioning

confidence: 99%

“…A reinforcement learning system mimics the way that humans learn. A reinforcement learning agent interacts with its environment in order to receive rewards [23]. This is in the same way that humans use their ears and eyes to perceive the world and act upon it.…”

Section: Reinforcement Learningmentioning

confidence: 99%

An Intelligent Framework for Estimating Software Development Projects using Machine Learning

Srivastava,

Agarwal

et al. 2023

IJRITCC

View full text Add to dashboard Cite

show abstract

“…A broad spectrum of techniques exists in the public domain to assess & estimate the effort required to test software [4], [5], [6], [7], [8], [9], [10], [11], [12], [13]. However, none of the literature cited includes any viable technique suitable to the early stages of commercial engagement.…”

Section: Client-side Issuesmentioning

confidence: 99%

An experimentally verified method to predict & measure all forms of Cyber-Risk

Storti¹

2021

Preprint

View full text Add to dashboard Cite

A precise & unambiguous mathematical definition of Cyber-Risk is developed, yielding an experimentally validated solution demonstrating ‘How to Predict & Measure Cyber-Risk’ for any Internet Connected Information System (ICIS) to greater than 98.07% accuracy. Moreover, it is shown that the solution holds for all scales of ICIS, from an Application level to an Enterprise level. In addition, it is shown that Test Effort Estimation (TEE) quantifies Cyber-Confidence, which in turn quantifies Cyber-Risk. Hence, TEE is a Mission Critical Activity (MCA) when formulating Cyber-Risk Management Strategies & may be utilised prior to project commencement, in-flight or post facto as an assessment &/or auditing tool. The TEE Model Construct developed is a statistical based methodology whereby the evaluations/decisions made, result in the contraction or expansion of the ‘z-Score’ associated with an infinite population of database records. The primary advantage of this approach is that very little information is required client-side at the engagement stage in order to produce peer acceptable estimates of the required test effort, & to accurately predict & measure the associated Cyber-Risk. This approach empowers clients & service providers to precisely define whatever level of Cyber-Risk is to be contractually delivered, capable of being absorbed, or prepared to be absorbed by consensus. With the aid of a decision table, estimators are able to articulate & convey to the appropriate authorities, various levels of Cyber-Risk commensurate with the available resources. The TEE Model Construct developed, presents an experimentally verified methodology, cognizant of commercial realities, yielding the following key advantages; (i) it requires minimal inputs, (ii) it has a scientific foundation, (iii) it facilitates operational decision-making, (iv) it quantifies Risk Based Testing (RBT), (v) it is simple, robust, flexible, consistent, reusable & transparent, (vi) it is capable of scaling a projected solution from a known solution, (vii) it embraces Continuous Improvement Processes (CIP’s), (viii) it confines perceptual subjectivity predominantly to three variables & (ix), it commercially exists as an off-the-shelf product. <br>

show abstract

“…For appropriate resource allocation and to prevent project delays and overruns, effort estimating is essential to project planning and management. A software project's success or failure can be significantly influenced by how accurately the effort is estimated [1]. A project's failure, cost overruns, and missed deadlines can all be caused by inaccurate estimation.…”

Section: [1] Introductionmentioning

confidence: 99%

Ensemble Machine Learning and Software Development Effort Estimation

Gora,

Sinha

2023

JAC

View full text Add to dashboard Cite

Effort estimation is a crucial component of software development that aids in re-source allocation and planning for project managers. While faulty estimation can lead to cost overruns and project failure, an accurate estimation can stop project delays and overruns. Increasingly used in recent years for software development effort estimation, ensemble approaches combine numerous models to increase pre-diction accuracy and stability. This study investigates various ensemble learning approaches and assesses how well they work on a dataset of actual software projects. The findings demonstrate that ensembles perform better than single models and can estimate effort with high accuracy. This work also offers insights into the parameters, such as the diversity of models and the quality of input information, that influence the performance of ensemble approaches. The results indicate that ensemble approaches, which can be implemented using a variety of techniques like bagging, boosting, and stacking, can be a feasible strategy for enhancing software development effort estimation. However, the success of these methods depends on the careful selection of models and characteristics.

show abstract

Software Effort Estimation Using Machine Learning Techniques

Cited by 9 publications

References 15 publications

An Intelligent Framework for Estimating Software Development Projects using Machine Learning

An Intelligent Framework for Estimating Software Development Projects using Machine Learning

An experimentally verified method to predict & measure all forms of Cyber-Risk

Ensemble Machine Learning and Software Development Effort Estimation

Contact Info

Product

Resources

About