文章摘要

孙坦,刘峥,崔运鹏,鲜国建,黄永文.融合知识组织与认知计算的新一代开放知识服务架构探析[J].中国图书馆学报,2019,45(3):38~48
融合知识组织与认知计算的新一代开放知识服务架构探析
Analysis and Design of A New Generation of Open Knowledge Service System Integrating Knowledge Organization and Cognitive Computing
投稿时间:2019-04-04  
DOI:
中文关键词: 知识组织  认知计算  知识应用  机器智能
英文关键词: Knowledge organization  Cognitive computing  Knowledge application  Machine intelligence
基金项目:本文系国家科技图书文献中心专项“下一代开放知识服务平台总体设计及关键技术研发”和中国农业科学院科技创新工程项目 “大规模RDF数据转换机理与存储研究”(编号:CAAS ASTIP-2016-AII)的研究成果之一
作者单位E-mail
孙坦 中国农业科学院农业信息研究所农业部农业大数据重点实验室北京 100081 suntan@caas.cn,suntan@caas.cn 
刘峥 中国科学院文献情报中心,北京 100190  
崔运鹏 中国农业科学院农业信息研究所农业部农业大数据重点实验室北京 100081  
鲜国建 中国农业科学院农业信息研究所农业部农业大数据重点实验室北京 100081  
黄永文 中国农业科学院农业信息研究所农业部农业大数据重点实验室北京 100081  
摘要点击次数: 381
全文下载次数: 0
中文摘要:
      信息过载现象导致发现并获取有用信息变得越来越困难,用户急切需要精准的知识发现和问题解答服务,通过对当前实现精准知识发现的主要技术方法进行分析,本研究分别采用基于传统知识组织构建方法和基于深度学习的方法,面向湿地领域进行语义知识组织体系的构建和精准发现实验。实验分析证明,传统知识组织方法无法单独支持特定主题的精准知识发现,尽管基于词向量的深度学习方法可以有效弥补传统知识组织系统的局限,但会受到计算语料规模和质量的限制。因此,本研究最终提出融合知识组织与认知计算的基本思路和体系框架,分析了融合方案所涉及的关键技术,这对于构建新一代开放知识服务系统具有重要指导意义。图3。参考文献15。
英文摘要:
Information explosion has made it difficult to acquire useful knowledge, and users are more likely to expect precise results or answers directly. This paper provides a survey to gain an overall view of the current research status in precise knowledge discovering focusing on methodologies, technologies, and most distinctive characteristics, and proposes possible solutions with regards to some existing problems and future research in this area.

The authors selected the field of wetland to make tests on semantic knowledge base construction and precise knowledge discovery. Two main approaches are adopted and compared experimentally: one is traditional Knowledge Organization System (KOS),and the other is deep learning.
The results indicate that traditional knowledge organization technique cannot support the precise knowledge of specific topics, while deep learning based on word vector can make some improvements on it, though the effect is limited by the scale and quality of computational corpus. Then the authors propose the framework and key technologies of integrated approach of KOS and cognitive computing.
The comparison and integration between new feature words obtained by deep learning and existing KOS is weak in this paper, and the verification of precise knowledge discovery should be made in practical application scenarios.
The results show that the integrated approach of KOS and cognitive computing can be expected to improve the precise knowledge discovery, and achieve high efficiency of knowledge calculation and high accuracy of intelligent question & answer.
As far as the authors know, this is the first study presenting the idea of a new generation of open knowledge service system. It shows the capabilities and limitations of current research in precise knowledge discovery and provides directions to researchers by pointing out that the integration of KOS and cognitive computing will effectively improve the low precision of machine algorithms led by lacking high quality big data and semantic knowledge base. In addition, the proposed approach may break through the deep cross boundary fusion and automatic knowledge acquisition of multi source heterogeneous data, and then make great improvements on transformations from unstructured literature data to structured and semantic knowledge networks. 3 figs. 15 refs.
下载全文   查看/发表评论  下载PDF阅读器