Abstract
Extending the work of Downing & Badar (2022), this paper presents the history of software, artificial intelligence, software quality assurance, and a software QA architecture called the Quality Assurance Machine (QAM). Using Design Science Research (DSR), the QAM was extended to support software containing ML models. Using descriptive statistics and hypothesis testing, this paper answers the research question: is QAM effective for assuring the quality of software containing ML? Future DSR efforts will add a Process Quality engine to the QAM to monitor and improve the processes used to create software and models.
Keywords
Quality assurance, traditional software, machine learning, design science, hypothesis testing