Title: Defect prediction in software using spiderhunt-based deep convolutional neural network classifier

Authors: M. Prashanthi; Chandra Mohan Miryala

Addresses: Jawaharlal Nehru Technological University Hyderabad (JNTUH), Kukatpally, Hyderabad-Telangana State, 500 085, India; CMR Engineering College, Kandlakoya, Hyderabad, 50140, India ' Department of Computer Science and Engineering, Jawaharlal Nehru Technological University Hyderabad (JNTUH), Kukatpally, Hyderabad-Telangana State, 500 085, India

Abstract: In this research, the defects in the software are predicted using the deep CNN classifier by effectively optimising the classifier using spiderhunt optimisation. The effective communication and hunting characteristics of the spiderhunt are employed for tuning the classifier that boosts the classifier performance. The proposed spiderhunt optimisation not only optimises the classifier but also plays a significant role in the feature selection for the extraction of necessary features that helps in defect prediction. The proposed spiderhunt optimisation achieved an improvement rate of 1.009%, 1.083%, 0.578%, and 1.01% in terms of accuracy, precision, recall, and F-measure and is proved to be quite efficient compared to state of art methods.

Keywords: deep learning; spiderhunt optimisation; software defect prediction; software engineering; quality.

DOI: 10.1504/IJNVO.2022.130947

International Journal of Networking and Virtual Organisations, 2022 Vol.27 No.4, pp.337 - 357

Received: 25 Jun 2022
Accepted: 09 Nov 2022
Published online: 14 May 2023 *

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