Standard Research Grant: Toxic Docs 2.0 Research Infrastructure Project

过去二十年见证了信息技术创新的真正爆炸式发展，这使得大规模、海量的文档数字化成为可能。Toxic Docs利用这些创新，为用户提供世界上最大的有毒物质（铅、石棉、二恶英、多氯联苯、聚氯乙烯、二氧化硅等）材料数据库。其中包括内部备忘录；未发表的科学研究；董事会分钟;还有公共关系文件。“有毒文档”中的几乎所有材料以前都是无法访问的，但在高风险的法律诉讼中被释放出来，然后由我们的数据库公开。用户包括学术研究人员、环境卫生科学家、记者和政策分析师。它向所有人免费开放。“有毒文档”的这一版本在五个方面建立在前一版本的基础上。首先，它继续扩大收藏。它正在处理另外数百万页关于石棉和滑石粉的文件；多氯联苯;矽肺病和定期收到更多的材料。其次，一个名为Toxic Tools的配套网站将为用户提供除了全文搜索之外的其他方法。其中包括一个命名实体识别工具，它可以找出常见的名称、组织和地点；查找与区域设置相关的文档的地理解析器；还有一个网络猜测器，可以推断文件中提到的人之间的关系；等等。第三，一个新的《信息自由法》（Freedom of Information Act， FOIA）部门将利用对公共记录申请的回应来建立我们的档案。最后，应用程序编程接口（API）将允许对使用纯文本版本的数据感兴趣的用户这样做。《有毒医生》提供了必要的原始数据，以构建关于21世纪环境健康风险新世界的新叙述。它的开放性允许用户对这些物质如何在环境中如此广泛地流通做出自己的客观判断。该奖项反映了美国国家科学基金会的法定使命，并通过使用基金会的知识价值和更广泛的影响审查标准进行评估，被认为值得支持。

The past two decades have seen a veritable explosion in information technology innovations that allow for large-scale, mass digitization of documents. Toxic Docs leverages these innovations to provide users with access to the world’s largest database of material on toxic substances (lead, asbestos, dioxin, PCBs, polyvinyl chloride, silica, and more). These include internal memos; unpublished scientific studies; boardroom minutes; and public relations documents. Almost all the material in Toxic Docs was previously inaccessible but has been unleashed in high-stakes legal litigation, then made public by our database. Users include academic researchers, environmental health scientists, journalists, and policy analysts. It is open to all and free of charge. This iteration of Toxic Docs builds on the previous one in five ways. First, it continues expanding the collection. It is processing millions of additional pages of documents on asbestos and talc; PCBs; and silicosis and regularly receives more material. Second, a companion site, Toxic Tools, will offer users additional ways of probing the material in addition to full-text search. These include a named entity recognition tool that pries out common names, organizations, and places; a geography parser that fishes out documents related to locales; and a network guesser that infers relationships among people mentioned in documents; among others. Third, a new Freedom of Information Act (FOIA) wing will use responses from public records requests to build our collections. Finally, an Application Programming Interface (API) will allow users interested in working with plain-text versions of our data to do so. Toxic Docs provides the raw data necessary to construct fresh narratives about the new world of 21st-century environmental health risk. Its openness allows users to make their own objective judgments on how these substances came to circulate so widely in the environment.This award reflects NSF's statutory mission and has been deemed worthy of support through evaluation using the Foundation's intellectual merit and broader impacts review criteria.