ICSE 2023 (series) / Artifact Evaluation /
Artifacts for Keyword Extraction From Specification Documents for Planning Security Mechanisms
This artifact contains the data used for evaluations as well as the data collection tools to extend the dataset for training the machine learning model for Keyword Extraction From Specification Documents for Planning Security Mechanisms. The artifact is both reusable and available. The packaged data collection tools include two web scraping tools created to extract data from CVE and Vendor websites. The packaged data contains 296,931 vulnerability reports of 52,110 products from 23,971 vendors extracted from CVE and over 3000 product documentation files extracted into plain text format from a number of vendor websites. The artifact is available for long-term archival at https://zenodo.org/record/7578926.