BioASQ and PubAnnotation : Using linked annotations in biomedical question answering

semanticscholar(2017)

引用 0|浏览5
暂无评分
摘要
Motivation: The motivation for this proposal is to extrinsically evaluate the resources available in PubAnnotation and investigate the potential of this repository as an external component of systems with direct real-world biomedical applications, in particular biomedical question answering. Approach: In this regard, we propose to adjust biomedical question answering systems to take advantage of linked annotations available in PubAnnotation. Those systems can be used to answer biomedical questions from benchmark datasets of the BioASQ challenge, with and without using additional annotations from PubAnnotation. Apart from annotations on relevant biomedical literature, we could also use annotations on benchmark questions through the same repository. Using the BioASQ evaluation infrastructure we can compare the performance of system versions using different resources. Therefore, we can assess the effect of using linked annotations in answering biomedical questions, corresponding to real information needs of biomedical experts.
更多
查看译文
AI 理解论文
溯源树
样例
生成溯源树,研究论文发展脉络
Chat Paper
正在生成论文摘要