基于科研知识图谱的研究侧写生成方法研究与设计

Research and Design of Research Profiling based on Scientific Knowledge Graph

针对科技文献爆炸式增长带来的信息获取挑战,本文开展基于科研知识图谱的研究侧写生成方法研究,综合运用文本挖掘、自然语言处理等智能技术深度融合领域知识和大规模文献信息,提出基于科研知识图谱的研究侧写系统设计方案,包括领域知识全景图、热点主题分析、重要文献推荐列表、文献发展脉络图、高影响力专家推荐、侧写文档生成与下载等服务功能模块,实现领域内主题结构、文献发展脉络、科研主体等核心内容的多角度挖掘和全景式揭示,提升大规模科技文献的知识发现水平。

Faced with the challenge of gaining access to scholarly contents as scientific literature and knowledge expand, this paper researches on a research profiling approach based on scientific knowledge graphs, which aims to achieve the deep fusion and thorough disclosure of scientific resources and domain knowledge by employing text mining and natural language processing techniques, among others. Furthermore, a scheme of research profiling based on scientific knowledge graph is designed, including function modules of overall graph view, important literature list, literature roadmap, hotness topics list, high-impact experts, and profile viewing and download, to realize multi-angle excavation and panoramic disclosure of core contents such as theme structure, literature development, and research subjects in specific domain, and improve the knowledge discovery of large-scale scientific literature.