Product Entity and Company-Product Relation Corpus

本数据集由德国人工智能研究中心创建，专注于商业产品命名实体识别与关系抽取。数据集包含152份英文网络和社交媒体文档，涵盖约131,000字，标注了2,191个组织和1,717个产品实体，以及379个公司提供产品的关系。数据集通过半自动标注过程构建，旨在解决非标准B2B产品及其关系的自动抽取问题，适用于供应链监控和市场研究等领域。

This dataset was developed by the German Research Center for Artificial Intelligence (DFKI), focusing on named entity recognition and relation extraction for commercial products. It includes 152 English web and social media documents, totaling approximately 131,000 words, with annotations for 2,191 organization entities, 1,717 product entities, and 379 relations describing companies providing their products. Constructed via a semi-automated annotation process, this dataset aims to address the automatic extraction of non-standard B2B products and their associated relations, and is applicable to fields such as supply chain monitoring and market research.