2023
1.
Nembhard, Fitzroy D; Slhoub, Khaled A; Carvalho, Marco M
An Agent-Based Approach Toward Smart Software Testing Conference
Proceedings of the Future Technologies Conference, vol. 814, Lecture Notes in Networks and Systems book series (LNNS) Springer Nature Switzerland, 2023, ISBN: 978-3-031-47451-4.
Abstract | Links | BibTeX | Tags: AOSE, NLP, software testing, virtual assistant
@conference{nokey,
title = {An Agent-Based Approach Toward Smart Software Testing},
author = {Fitzroy D Nembhard and Khaled A Slhoub and Marco M Carvalho},
editor = {K. Arai},
doi = {https://doi.org/10.1007/978-3-031-47451-4_21},
isbn = {978-3-031-47451-4},
year = {2023},
date = {2023-11-01},
urldate = {2023-11-01},
booktitle = {Proceedings of the Future Technologies Conference},
volume = {814},
publisher = {Springer Nature Switzerland},
series = { Lecture Notes in Networks and Systems book series (LNNS)},
abstract = {As the field of software testing continues to advance, cooperative software testing and analysis has been proposed as a new methodology to help combat the challenges involved with performing highly effective software testing. Motivated by the understanding that auto-testing systems on their own are not powerful enough to address complications in testing complex real-world software, this model involves human-machine and machine-machine cooperation to make automated software testing processes more interactive and user-friendly. It is with this in mind that we propose an agent-based approach to software testing that involves teaming humans with virtual assistants on smart devices to help coordinate the tasks associated with testing complex real-world software. Currently, virtual assistants are widely used for interpersonal tasks such as purchasing items from restaurants, interfacing with mobile applications to create events and reminders, and composing and sending messages on behalf of the user. In this research, we create an agent-based framework and use it to demonstrate that a virtual assistant on a smart device can also be utilized to work closely with software testers to efficiently and effectively verify software functionality, generate reports, and communicate results with developers. We utilize unit testing to evaluate our proposed methodology by applying it to a set of Java projects. Our results show that virtual agents can be used to work with humans to coordinate tasks associated with unit-testing software.},
keywords = {AOSE, NLP, software testing, virtual assistant},
pubstate = {published},
tppubtype = {conference}
}
As the field of software testing continues to advance, cooperative software testing and analysis has been proposed as a new methodology to help combat the challenges involved with performing highly effective software testing. Motivated by the understanding that auto-testing systems on their own are not powerful enough to address complications in testing complex real-world software, this model involves human-machine and machine-machine cooperation to make automated software testing processes more interactive and user-friendly. It is with this in mind that we propose an agent-based approach to software testing that involves teaming humans with virtual assistants on smart devices to help coordinate the tasks associated with testing complex real-world software. Currently, virtual assistants are widely used for interpersonal tasks such as purchasing items from restaurants, interfacing with mobile applications to create events and reminders, and composing and sending messages on behalf of the user. In this research, we create an agent-based framework and use it to demonstrate that a virtual assistant on a smart device can also be utilized to work closely with software testers to efficiently and effectively verify software functionality, generate reports, and communicate results with developers. We utilize unit testing to evaluate our proposed methodology by applying it to a set of Java projects. Our results show that virtual agents can be used to work with humans to coordinate tasks associated with unit-testing software.