Within-Project Defect Prediction of Infrastructure-as-Code Using Product and Process Metrics

Infrastructure-as-code (IaC) is the DevOps practice enabling management and provisioning of infrastructure through definition machine-readable files, hereinafter referred to as IaC scripts . Similarly other source code artefacts, these files may contain defects that can preclude their correct functioning. In this paper, we aim at assessing role xmlns:xlink="http://www.w3.org/1999/xlink">product xmlns:xlink="http://www.w3.org/1999/xlink">process metrics when predicting defective IaC scripts. We propose a fully integrated machine-learning framework for Defect Prediction, allows repository crawling, collection, model building, evaluation. To evaluate it, analyzed 104 projects employed five classifiers compare performance in flagging suspicious The key results study report Random Forest best-performing model, with median AUC-PR 0.93 MCC 0.80. Furthermore, least collected projects, product identify scripts more accurately than process metrics. Our findings put baseline investigating Prediction relationship between metrics, scripts’ quality.