基于复杂关联网络的生物医学研究结构的挖掘

摘要

A biomedical entity association evolution network was constructed by mining the implicit associations in PubMed-covered literature, which can help scientific researchers to form new scientific hypotheses, to analyze the topological features of associated network, to study the scientific literature-enriched knowledge structure, associa-tions, development rules, to introduce new visual angles and methods for literature-based knowledge discovery, and to improve the knowledge discovery efficiency.%基于免费开放的PubMed文献数据集，利用文献的知识发现，通过挖掘文献中隐含的关联，构建了生物医学实体关联演化网络。它能帮助科研人员形成新的科学假设，分析关联网络的拓扑特征，从系统层面上研究科学文献富集的知识结构、相关性与发展规律，为文献的知识发现引入新的视角与方法，提高知识发现的效率。