Expert-Annotated Dataset for Bug Severity and Priority Prediction in Software Development

This dataset contains a representative subset of 200 anonymized bug report records used in the research paper “Machine Learning Model for Accurate Bug Severity and Priority Prediction in Software Development.” The complete dataset comprises 2,000 industry-collected UI bug reports, manually labeled by a panel of domain experts including QA engineers, senior developers, and academic specialists. Each record is represented by 15 defined features and two target labels: Severity (Critical, Major, Minor) and Priority (High, Medium, Low). This dataset aims to support reproducibility, benchmarking, and further research in automated bug triaging, defect management, and software quality assurance using machine learning techniques.