Data analytics for "cost of quality" of code artefacts.

点击翻译按钮获取中文摘要

With the increasing size and complexity of software systems, more and more adaptive ways have been proposed to accommodate the high demands not only for correctness, but also performance, security, reliability and other quality criteria. With the shortage of personal, investing into developers' skills by enabling them to identify and fix their own coding flaws is considered valuable. The more this process can be facilitated and partially automated, the higher its return of investment for CodeExellence. Expected cost savings are based on (i) improved time to market, while improving customer satisfaction, (ii) fixing critical defects in code while it is still in development, when it is most cost and time effective to fix them, (iii) ensures good development practices that result in components and frameworks that can be reused, thereby reducing rework and improving delivery times, and (iv) drives application maintenance costs down through automated and continuous structural quality improvement. The ENGAGE project is intended to leverage expertise in artificial intelligence, data analytics, release engineering and software engineering decision support for the sake of automating the control and diagnostic process in software. The pathway to achieve this collaboration project objective is to analyze, design and evaluate big data available from existing repositories. As a form of diagnostic analytics, the intention is to extract trends and patterns inherent in the data. These trends are planned to guide and automate the decision-making process on "Which bugs should be fixed?" More precisely, the research will be on (i) Methods to extract patterns, (ii) Characterize patterns, and (iii) Evaluate the applicability and usefulness of these patterns.