摘要
数字图书馆知识发现并非是单一线索下的线性知识延展，相反却是极具多维性且在实践工作中不同维度之间总是相互关联、相互作用的。目前，数字图书馆资源聚合与知识发现的途径主要通过组织结构维度下的领域本体来实现，其语义化程度较高但却无法摆脱概念树结构的桎梏。而关联关系维度则能够突破主题领域内部严谨苛刻的层级限制，基于知识问的关联关系将平面树形结构拓展到网络立体空间。传递扩散维度更是将知识的客观属性与人类的主观认知相结合，充分体现了知识与人的交互。笔者认为，以多维度融合的视角考查数字图书馆资源聚合与知识发现，最大程度上弥合了学术界当前存在的分歧，使基于数字图书馆资源聚合的知识发现成为可能，有助于突破数字图书馆知识组织研究领域的瓶颈，把数字图书馆资源聚合与知识发现研究推向一个新的高度。
Digital library knowledge discovery is not linear extension of knowledge under single clue but is multi- dimensional, different dimensions are always interrelated and interacting in practice. At present, the approach of digital library resource aggregation and knowledge discovery is realized through the ontology under the dimension of organizational structure, which is in high degree of semantization but cannot get rid of the shackles of the concept-tree structure. However, dimension of relevance can break the rigorous and demanding hierarchy restrictions inside topic areas. Relationship based on knowledge spread surface tree structure into network three-dimensional space. Dimension of spread combines knowledge＇ s objective attribute with people＇ s subjective perception, reflecting the interaction between knowledge and human. We believe, analyzing the digital library resource aggregation and knowledge discovery in perspective of multi- dimensional integration can bridge current disagreements in academia to the maximum extent, which makes possible knowledge discovery based on digital library resource aggregation, can help to break the bottleneck in the field of digital library knowledge organization, and can also pushes the digital library resource aggregation and knowledge discovery research to a new level.
出处
《情报学报》
CSSCI
北大核心
2014年第2期148-157,共10页
Journal of the China Society for Scientific and Technical Information
基金
国家自然科学基金项目“语义网络环境下数字图书馆资源多维度聚合与可视化展示研究”(编号:71273111)
国家社会科学基金重大项目“基于语义的馆藏资源深度聚合与可视化研究”(项目编号:11&ZD152)
吉林大学985工程项目的研究成果之一
关键词
数字图书馆
知识发现
组织结构维度
传递扩散维度
关联关系维度
digital library
knowledge discovery
dimension of organizational structure
dimension of spread
dimensionof relevance
