A Novel Hybrid Algorithm for Software Cost Estimation Based on Cuckoo Optimization and K-Nearest Neighbors Algorithms

Journal Title: Engineering, Technology & Applied Science Research - Year 2016, Vol 6, Issue 3

Abstract

The inherent uncertainty to factors such as technology and creativity in evolving software development is a major challenge for the management of software projects. To address these challenges the project manager, in addition to examining the project progress, may cope with problems such as increased operating costs, lack of resources, and lack of implementation of key activities to better plan the project. Software Cost Estimation (SCE) models do not fully cover new approaches. And this lack of coverage is causing problems in the consumer and producer ends. In order to avoid these problems, many methods have already been proposed. Model-based methods are the most familiar solving technique. But it should be noted that model-based methods use a single formula and constant values, and these methods are not responsive to the increasing developments in the field of software engineering. Accordingly, researchers have tried to solve the problem of SCE using machine learning algorithms, data mining algorithms, and artificial neural networks. In this paper, a hybrid algorithm that combines COA-Cuckoo optimization and K-Nearest Neighbors (KNN) algorithms is used. The so-called composition algorithm runs on six different data sets and is evaluated based on eight evaluation criteria. The results show an improved accuracy of estimated cost.

Authors and Affiliations

E. E. Miandoab, F. S. Gharehchopogh

Keywords

Related Articles

Enhancement of Edge-based Image Quality Measures Using Entropy for Histogram Equalization-based Contrast Enhancement Techniques

An Edge-based image quality measure (IQM) technique for the assessment of histogram equalization (HE)-based contrast enhancement techniques has been proposed that outperforms the Absolute Mean Brightness Error (AMBE) and...

Assessment of the Suitability of Paper Waste as an Engineering Material

This research work investigates the potential applicability of waste paper in the production of ceiling boards with focus on achieving: environmental sustainability, safe disposal of waste paper and more cost effective p...

Growth mode study of MgCl2 on Au foil and Si (111) 7x7, under Ultra High Vacuum by XPS

The growth mode of MgCl2 on Au foil and Si (111) 7x7 reconstructed surface under UHV conditions, was investigated by X-ray Photoelectron Spectroscopy (XPS). Magnesium chloride grows with the Frank-van der Merve, (FM) gro...

A Review on Self Compacting Concrete with Cementitious Materials and Fibers

Self-compacted concrete (SCC) is cast in the formwork without compaction and it fulfills the formwork due to its own weight. SCC is considered to have many advantages in comparison with conventional concrete like improve...

Numerical Study of the Thermal Behavior of a Composite Phase Change Material (PCM) Room

In this study, thermal performance of building walls integrated with phase change materials (PCM) was evaluated in terms of indoor temperature reduction and heat transfer time delay. PCM was incorporated as thin layer pl...

Download PDF file
  • EP ID EP109468
  • DOI -
  • Views 245
  • Downloads 0

How To Cite

E. E. Miandoab, F. S. Gharehchopogh (2016). A Novel Hybrid Algorithm for Software Cost Estimation Based on Cuckoo Optimization and K-Nearest Neighbors Algorithms. Engineering, Technology & Applied Science Research, 6(3), -. https://europub.co.uk/articles/-A-109468