当前位置: X-MOL 学术International Journal on Digital Libraries › 论文详情
Our official English website, www.x-mol.net, welcomes your feedback! (Note: you will need to create a separate account there.)
Assessing plausibility of scientific claims to support high-quality content in digital collections
International Journal on Digital Libraries Pub Date : 2018-10-28 , DOI: 10.1007/s00799-018-0256-8
José María González Pinto , Wolf-Tilo Balke

This paper presents a formalization and extension of a novel approach to support high-quality content in digital libraries. Building on the concept of plausibility used in cognitive sciences, we aim at judging the plausibility of new scientific papers in light of prior knowledge. In particular, our work proposes a novel assessment of scientific papers to qualitatively support the work of reviewers. To do this, our approach focuses on the key aspect of scientific papers: claims. Claims are sentences found in empirical scientific papers that state statistical associations between entities and correspond to the core contributions of the papers. We can find these types of claims, for instance, in medicine, chemistry, and biology, where the consumption of a drug, a substance, or a product causes an effect on some other type of entity such as a disease, or another drug or substance. To operationalize the notion of plausibility, we promote claims as first-class citizens for scientific digital libraries and exploit state-of-the-art neural embedding representations of text and topic models. As a proof of concept of the potential usefulness of this notion of plausibility, we study and report extensive experiments on documents with scientific papers from the PubMed digital library.

中文翻译:

评估科学主张的合理性以支持数字馆藏中的高质量内容

本文介绍了一种新方法的形式化和扩展,以支持数字图书馆中的高质量内容。在认知科学中使用合理性概念的基础上,我们旨在根据先验知识判断新科学论文的合理性。特别是,我们的工作提出了对科学论文的新颖评估,以定性地支持审稿人的工作。为此,我们的方法侧重于科学论文的关键方面:主张。声明是在经验科学论文中发现的句子,陈述了实体之间的统计关联,并与论文的核心贡献相对应。我们可以在医学,化学和生物学等领域找到这些类型的声明,其中消耗某种药物,一种物质或一种产品会对某种其他类型的实体(例如疾病,或其他药物或物质。为了使合理性的概念付诸实践,我们以科学数字图书馆的一等公民身份倡导主张,并利用文本和主题模型的最新神经嵌入表示法。为了证明这种合理性概念可能有用的概念,我们使用PubMed数字图书馆的科学论文研究并报告了关于文档的广泛实验。
更新日期:2018-10-28
down
wechat
bug