Fighting COVID-19 with computational tools: an AI guided review of 17,000 studies - The CSCoV database.
收藏资源简介：
CSCoV (Computational Studies about COVID-19) is a dataset containing COVID-19 related studies extracted from PubMed, bioRxiv, medRxiv, and arXiv, together with article and author related metrics obtained from Semantic Scholar (plus page views from bioRxiv and medRxiv). Using machine learning, the articles are categorized in six topics (Pharmacology, Genomics, Epidemiology, Healthcare, Clinical Medicine, Clinical Imaging) and prioritized. The database is periodically updated. Publication: TBA Files included in this release: cscov_07_2021.png: dataset statistics for the current CSCoV release. cscov_07_2021.tsv: CSCoV database. schema.json: metadata. cscov_07_2021.tar.gz: Doc2Vec and DeepWalk features used for the DL model Source code: https://github.com/SFB-KAUST/covid-review
CSCoV（Computational Studies about COVID-19，新冠计算研究数据集）是一款收录了从PubMed、bioRxiv、medRxiv及arXiv平台提取的新冠相关研究文献的数据集，同时附带从Semantic Scholar获取的文章与作者相关指标，并补充了bioRxiv与medRxiv的页面浏览量数据。研究人员通过机器学习技术将上述文献划分为六大主题，分别为药理学、基因组学、流行病学、医疗保健、临床医学与临床影像学，并对其进行优先级排序。该数据库将定期进行更新。出版信息：待定（TBA）。本发布版本包含的文件如下：cscov_07_2021.png：当前CSCoV版本的数据集统计信息图表；cscov_07_2021.tsv：CSCoV数据库本体文件；schema.json：元数据文件；cscov_07_2021.tar.gz：深度学习模型所用的Doc2Vec与DeepWalk特征文件。源代码链接：https://github.com/SFB-KAUST/covid-review
