Title: Extended COCOMO: robust and interpretable neuro-fuzzy modelling

Authors: Brajesh Kumar Singh; Shailesh Tiwari; Krishn Kumar Mishra; Akash Punhani

Addresses: Department of Computer Science and Engineering, Raja Balwant Singh Engineering Technical Campus, Agra, India ' Department of Computer Science and Engineering, ABES Engineering College, Ghaziabad, India ' Department of Computer Science and Engineering, MNNIT Allahabad, Prayag, India ' Department of Computer Science and Engineering, ASET, Amity University, Noida, India

Abstract: Prediction of software development efforts is one of the crucial activities in software project management. Still, search for the perfect model for software cost estimation has become most difficult task of the organisations dealing in software development. This paper presents the extended version of COCOMO, which is done with the help of two very popular methods, i.e., artificial neural networks (ANN) and fuzzy logic. Firstly, the expert judgement about model is used for validation, and overpowers the common software engineering 'black box' problem that arises widely in ANN-based solutions. Moreover, we choose the best combination of one of the three membership functions for continuous-rating values, which reduce the variance while estimating the cost of similar projects. The validation, using 93 NASA projects dataset, shows that the model significantly improves the estimation accuracy in terms of mean magnitude of relative error (MMRE) by 10.104% relative to other known estimation models.

Keywords: fuzzy logic; neural network; constructive cost model; COCOMO; neuro-fuzzy software effort estimation; NASA projects dataset; mean magnitude of relative error; MMRE.

DOI: 10.1504/IJCVR.2021.111873

International Journal of Computational Vision and Robotics, 2021 Vol.11 No.1, pp.41 - 65

Received: 07 Jan 2019
Accepted: 02 Jul 2019
Published online: 18 Dec 2020 *

Full-text access for editors Full-text access for subscribers Purchase this article Comment on this article