Dataset of duplicate vulnerability records across databases

This dataset contains vulnerability duplicate information from two sources: the cross-database duplicates and the GitHub Advisory Database duplicates. The dataset is provided in JSON format and is intended for use in research related to vulnerability matching and duplication detection. ## Dataset Overview The dataset consists of two files: 1. **cross_database_duplicates.json**: Contains 22,163 pairs of duplicate vulnerabilities identified across multiple databases.2. **github_advisory_database_duplicates.json**: Contains 133 pairs of duplicate vulnerabilities specifically from the GitHub Advisory Database. ## File Format Both files are in JSON format. Each record consists of four attributes: - `id_1`: The ID of the first vulnerability report.- `id_2`: The ID of the second vulnerability report.- `record_1`: The first vulnerability report.- `record_2`: The second vulnerability report. These attributes are designed to help users identify and compare vulnerability reports that are considered duplicates. ## Usage This dataset can be used for studies in vulnerability matching, natural language processing (NLP) applications, and the development of tools for detecting duplicate vulnerabilities in different databases.