Title: Ensemble approach of GP, ACOT, PSO, and SNN for predicting software reliability

Authors: D. Shanthi; Narla Swapna; Ajmeera Kiran; Shaga Anoosha

Addresses: California Public University, Delaware, USA; Department of Information Technology, Maturi Venkata Subba Rao (MVSR) Engineering College, Hyderabad, Telangana, India ' Department of Computer Science and Engineering, CMR College of Engineering and Technology, Telangana, India ' Department of Computer Science and Engineering, MLR Institute of Technology (MLRIT), Telangana, India ' Department of Computer Science and Engineering, Vardhaman College of Engineering, Hyderabad, Telangana, India

Abstract: In recent decades, software has grown in importance. More and more computing systems are being intefted into modern society, increasing the necessity for rigorous software development methodologies. Software crises are issues that create delays, increased expenses, or failure to meet user needs. This difficult endeavour can be made easier by enhancing the software development process. We proposed GP, ACOT, PSO, SNN, and a mixture of GP, ACOT, PSO, and SNN to predict software reliability. Our results were compared to existing machine learning algorithms like neural networks and decision trees. We collected three software failure datasets using RMSE and NRMSE to support the need.

Keywords: decision tree; ant colony optimisation techniques; particle swarm optimisation techniques; spiking neural networks; soft computing techniques; software reliability; machine learning.

DOI: 10.1504/IJESMS.2024.136976

International Journal of Engineering Systems Modelling and Simulation, 2024 Vol.15 No.2, pp.68 - 75

Received: 13 Nov 2021
Accepted: 28 Jan 2022
Published online: 01 Mar 2024 *

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