| 基于多源数据知识融合方法的研究前沿识别 |
| |
| 引用本文： | 谭晓,李辉.基于多源数据知识融合方法的研究前沿识别[J].现代情报,2019,39(8):29-36. |
| |
| 作者姓名： | 谭晓 李辉 |
| |
| 作者单位： | 北京市科学技术情报研究所, 北京 100044 |
| |
| 基金项目： | 国家社会科学基金项目"总体国家安全观视阈下的科技情报服务创新研究"（项目编号：18BTQ054）。 |
| |
| 摘 要： | 目的/意义]面对科技创新演变的加剧和交叉融合加速的大环境,利用情报研究方法及其他学科方法准确识别科技前沿成为获取科技战略情报的重要任务。研究前沿不仅提供了对当前重点和未来趋势的预见,而且为政府决策提供了关键指标。方法/过程]通过内容分析对当前研究前沿的识别框架、方法以及多元关系、深入内容层面分析方法的应用等现状进行总结,发现目前在研究前沿识别模型和方法中仍存在不足。结果/结论]针对不足,结合多源数据进行知识融合初步设计了综合宏观和微观的前沿识别模型,将多实体和多关系融合应用到主题关联,利用图模型的社团结构识别和Clique所含信息进行主题表示;划分研究前沿类型并构建前瞻性指标体系,完成科技前沿的识别,以期更准确、高效、全面地识别科技前沿。
|
| 关 键 词： | 研究前沿 多源数据 知识融合方法 多元关系 主题识别 知识内容 |
Research Front Recognition Based on Multi-source Data Knowledge Fusion Method |
| |
| Authors: | Tan Xiao Li Hui |
| |
| Institution: | Beijing Institute of Science and Technology Information, Beijing 100044, China |
| |
| Abstract: | Purpose/Significance]In the context of accelerating evolution of scientific and technological innovation and accelerating cross-integration,it is an important task to accurately identify the front of science and technology by using intelligence research methods and other disciplinary methods to obtain strategic information of science and technology.Research front not only provide predictions of current priorities and future trends,but also provide key indicators for government decision-making.Method/Process]Through content analysis,this paper summarized the current status of the research front identification framework,methods and multiple relations of the current research front,as well as the application of in-depth analysis methods at the content level,and found that there were still deficiencies in the current research front identification models and methods.Result/Conclusion]In view of the shortcomings,a comprehensive macro and micro front identification model was preliminarily designed by combining knowledge fusion of multi-source data.Multi-entity and multi-relation1 fusion were applied to subject association,and community structure identification of graph model and clique information were used for subject representation.Divide research front types and construct forward-looking index system to complete the identification of science and technology front,so as to identify science and technology front more accurately,efficiently and comprehensively. |
| |
| Keywords: | research front multi-source data knowledge fusion method multiple relations topic detection knowledge content |
| 本文献已被 维普 等数据库收录！ |
| 点击此处可从《现代情报》浏览原始摘要信息 |
| 点击此处可从《现代情报》下载免费的PDF全文 |
|
