Information Technology Group, Wageningen University & Research, Wageningen, The Netherlands.
Department of Primary and Community Care, Radboud University Medical Center, Nijmegen, The Netherlands.
J Intellect Disabil Res. 2020 Jul;64(7):475-481. doi: 10.1111/jir.12730. Epub 2020 Apr 27.
Corona virus disease 2019 (COVID-19) has been announced as a new coronavirus disease by the World Health Organization. At the time of writing this article (April 2020), the world is drastically influenced by the COVID-19. Recently, the COVID-19 Open Research Dataset (CORD-19) was published. For researchers on ID such as ourselves, it is of key interest to learn whether this open research dataset may be used to investigate the virus and its consequences for people with an ID.
From CORD-19, we identified full-text articles containing terms related to the ID care and applied a text mining technique, specifically the term frequency-inverse document frequency analysis in combination with K-means clustering.
Two hundred fifty-nine articles contained one or more of our specified terms related to ID. We were able to cluster these articles related to ID into five clusters on different topics, namely: mental health, viral diseases, diagnoses and treatments, maternal care and paediatrics, and genetics.
The CORD-19 open research dataset consists of valuable information about not only COVID-19 disease but also ID and the relationship between them. We suggest researchers investigate literature-based discovery approaches on the CORD-19 and develop a new dataset that addresses the intersection of these two fields for further research.
世界卫生组织宣布 2019 年冠状病毒病(COVID-19)为一种新型冠状病毒病。在撰写本文时(2020 年 4 月),全世界正受到 COVID-19 的巨大影响。最近,COVID-19 开放研究数据集(CORD-19)发布。对于我们这样的 ID 研究人员来说,关键是要了解这个开放的研究数据集是否可用于研究该病毒及其对 ID 患者的影响。
我们从 CORD-19 中识别出包含与 ID 护理相关术语的全文文章,并应用了一种文本挖掘技术,特别是术语频率-逆文档频率分析与 K-均值聚类相结合的方法。
有 259 篇文章包含一个或多个我们指定的与 ID 相关的术语。我们能够将这些与 ID 相关的文章聚类为五个不同主题的聚类,即:心理健康、病毒疾病、诊断和治疗、母婴保健和儿科以及遗传学。
CORD-19 开放研究数据集不仅包含有关 COVID-19 疾病的有价值信息,还包含有关 ID 及其之间关系的信息。我们建议研究人员对 CORD-19 进行基于文献的发现方法研究,并开发一个新的数据集,以进一步研究解决这两个领域交叉的问题。