FAIR Requirements-Dataset

This record provides a consolidated and FAIR-oriented dataset of technical product requirements created for research and teaching in Requirements Engineering. The dataset integrates 11 heterogeneous sources and contains 7841 requirements in German and English. Each requirement is represented as a uniquely identifiable entry and enriched with core context metadata such as project reference, domain assignment, maturity level, licensing information where available, and a source URL field for provenance tracking. The release is provided in two complementary target formats: JSON Lines for data-driven workflows and ReqIF for interoperability with requirements management tools. The package includes only the final Dataset, without the curated folder structure with category-specific subsets (Categories A–C). A README file documents the data model, label sets, and reuse guidance.Intended use: benchmarking, NLP/ML experiments (e.g., classification, quality checks, traceability support), as well as corpus-based analyses of technical product requirements across domains.