Scientific Literature Recommendation and Knowledge Extraction for Emergency Needs
[Purpose/significance]To meet the public information needs during public emergencies,a model is proposed to recommend scientific and technical literature and extract knowledge from the literature.The model is able to recommend high quality literature and scientific and technical knowledge relevant to public emergency needs to the public.[Method/process]Firstly,the microblog posts related to"Yang Kang"(recovery from Covid-19)and"Sequelae"on the Sina Weibo platform are clustered to mine the public emergency needs of the Covid-19 outbreak.Secondly,we recommend high-quality literature.Topic keywords are extracted from the literature.The similarity between keywords and public emergency needs is computed,and the literatures with the similarity higher than the threshold are selected as candidate recommended literature collection.Indicators of similarity,innovation,academic influence and Altmetric are considered to rank the candidate literatures.The rule-based approach is used to extract the knowledge in the litera-tures.[Result/conclusion]The public emergency needs are summarized into six categories,which are mild symptoms,post-rehabilitation exercise,protective measures,secondary infection,mental health,and nutrition management.[Innovation/limitation]This study recommends high-quality literature and technical knowledge related to the public needs to social media users.In future,we will investigate more types of public emergencies,mine public emergency needs during different events and recommend related litera-tures.
emergency needsliterature recommendationscientific and technical knowledgeknowledge extractioninformation matching
NETL
NSTL
万方数据
