Examination of software testing activities and softwarequality estimation with artificial intelligence techniques

This thesis emphasizes the decisive role of software testing processes in enhancing software quality and aims to demonstrate how these processes can be managed more efficiently, predictably, and strategically through artificial intelligence and machine learning techniques. With the accelerating pace of digitalization, not only functionality but also quality factors such as reliability, sustainability, and user satisfaction have gained prominence in software projects, making the planning and execution of testing processes a key determinant of software quality. To overcome the limitations of traditional methods, this study applied various machine learning algorithms to survey data collected from software test engineers working in the field, and developed classification models to predict the quality level of software testing processes. The findings reveal that the proposed models generate reliable outputs that can support decisionmakers in managing software test processes, demonstrating that this approach can make concrete contributions to both the academic literature and sectoral practice.