Research Article
A New Method for Enhancing Software Effort Estimation by Using ANFIS-Based Approach
@INPROCEEDINGS{10.1007/978-3-030-77424-0_16, author={The-Anh Le and Quyet-Thang Huynh and Tran-Tuan-Nam Nguyen and Minh-Hoa Tran Thi}, title={A New Method for Enhancing Software Effort Estimation by Using ANFIS-Based Approach}, proceedings={Industrial Networks and Intelligent Systems. 7th EAI International Conference, INISCOM 2021, Hanoi, Vietnam, April 22-23, 2021, Proceedings}, proceedings_a={INISCOM}, year={2021}, month={5}, keywords={ALBRE Effort estimation ANFIS Function point FPA Software estimation Neural-fuzzy}, doi={10.1007/978-3-030-77424-0_16} }- The-Anh Le
Quyet-Thang Huynh
Tran-Tuan-Nam Nguyen
Minh-Hoa Tran Thi
Year: 2021
A New Method for Enhancing Software Effort Estimation by Using ANFIS-Based Approach
INISCOM
Springer
DOI: 10.1007/978-3-030-77424-0_16
Abstract
The accurate estimation of the effort and cost becomes one of the important issues of project management. There are some algorithmic and non-algorithmic techniques are already developed to tackle the challenges of estimation tools in software project management such as Bayes probability-based approach, classification and regression, semantic analysis of software requirements, artificial neural networks, fuzzy logic, and hybrid methods. These techniques are unable to satisfy the management of modern and dynamic software development process. The aim of this paper is to propose a method of estimation by using the fuzzy logic’s related functions and the fuzzy algorithm of the Adaptive Neuro-Fuzzy Inference System (ANFIS) model to improve the accuracy of Functional Point Analysis (FPA) to estimate the cost and effort of software development. A tool called ALBRE is also developed to support the calculation of the proposed method. Experimental results show that the proposed method based on the ANFIS model produces positive results, with less errors. The accuracy of VAF increases by up to 80% compared to the method proposed by Albrecht in Function Point Counting Practices Manual 4.2.1 by considering the Mean Magnitude of Relative Error (MMRE).
