Robotic Automation of Software Testing From a Machine Learning Viewpoint

Abstract

The need to scale software test automation while managing the test automation process within a reasonable time frame remains a crucial challenge for software development teams (DevOps). Unlike hardware, the software cannot wear out but can fail to satisfy the functional requirements it is supposed to meet due to the defects observed during system operation. In this era of big data, DevOps teams can deliver better and efficient code by utilizing machine learning (ML) to scan their new codes and identify test coverage gaps. While still in its infancy, the inclusion of ML in software testing is a reality and requirement for coming industry demands. This study introduces the prospects of robot testing and machine learning to manage the test automation process to guarantee software reliability and quality within a reasonable timeframe. Although this paper does not provide any particular demonstration of ML-based technique and numerical results from ML-based algorithms, it describes the motivation, possibilities, tools, components, and examples required for understanding and implementing the robot test automation process approach.