摘要
【目的/意义】为了维护南海领土主权权益和领土完整,同时为科研人员或相关部门提供充足的南海维权档案资料,辅助他们的科研工作,提高科研效率。【方法/过程】本文首先通过CiteSpace详细论述目前国内外南海档案及南海档案系统研究现状;其次,运用文献调研法,整理多源异构的南海档案资料,由于工作量艰巨,本文最终从南海文库数字资源中随机抽取302篇南海档案资料作为本文的数据基础;再次,应用正则表达式从多模态南海档案资料中抽取具体的一句话作为细粒度南海档案数据;为了进一步提取细粒度南海维权档案数据,通过人工标注2272条有价值和没有价值的细粒度南海档案数据,应用机器学习和深度学习算法对细粒度南海档案数据进行抽取及比较;最后应用知识图谱构建南海维权档案系统,实现细粒度南海档案数据的检索及可视化。【结果/结论】本文最终选择DNN模型抽取细粒度南海档案数据,其准确率达0.86。本研究具有一定的可行性,可以为后续的研究者抛砖引玉。【创新/局限】面向总体国家安全观构建南海维权档案系统,初步实现细粒度南海维权档案数据检索及可视化。
[Purpose/significance]In order to safeguard territorial sovereignty and integrity in the South China Sea and provide ample archives and information on South China Sea rights protection for researchers and relevant institutions,thus assisting them in their sci-entific research work and improving research efficiency.[Method/process]This article first provides a detailed exposition of the cur-rent research status of South China Sea archives and South China Sea archive systems both domestically and internationally using CiteSpace.Secondly,through literature research,it organizes diverse and heterogeneous South China Sea archival materials.Due to the enormous workload,this article randomly selects 302 South China Sea archival materials from the digital resources of the South China Sea Library as the data foundation.Thirdly,by applying regular expressions,it extracts specific sentences as fine-grained South China Sea archival data from multi-modal South China Sea archival materials.In order to further extract fine-grained South China Sea rights protection archival data,2272 valuable and non-valuable fine-grained South China Sea archival data are manually annotated,and machine learning and deep learning algorithms are applied to extract and compare the fine-grained South China Sea archival data.Finally,a knowledge graph is applied to construct the South China Sea rights protection archives system,enabling the retrieval and vi-sualization of fine-grained South China Sea archival data.[Result/conclusion]In the end,this article selects the DNN model to extract fine-grained South China Sea archival data,achieving an accuracy of O.86.This research demonstrates a certain feasibility and can serve as a starting point for future researchers.[Innovation/limitation]Build the South China Sea Rights Protection Archives System fo-cusing on the overall national security concept,and preliminarily realize the fine-grained retrieval and visualization of the South China Sea Rights Protection Archives data.
作者
彭玉芳
王妍睿
杨海平
徐浩
孙宁
PENG Yufang;WANG Yanrui;YANG Haiping;XU Hao;SUN Ning(School of Economics&Management,Nanjing Institute of technology,Nanjing 211167,China;Department of Information Management,Nanjing University,Nanjing 210046,China)
出处
《情报科学》
CSSCI
北大核心
2024年第5期35-47,共13页
Information Science
基金
教育部人文社会科学基金青年资助项目“基于多模态深度学习的细粒度南海疆证据资料抽取与分类研究”(22YJC870012)
江苏省社科基金后期资助项目“南海维权证据链构建与智库服务研究”(22HQB43)
江苏省高校哲学社会科学研究一般项目“南海疆维权证据链的元数据组织模式研究”(2022SJYB0444)
南京工程学院校级科研基金“南海维权证据链智能检索系统构建研究”(YKJ202231)
国家社会科学基金重大项目“南海疆文献资料整理中的知识发现与维权证据链建构研究”(19ZDA347)
江苏省社科基金青年项目“融合在线学术社交信息的电子文献资源推荐策略研究”(21TQC003)。
关键词
南海维权档案系统
细粒度南海档案数据
信息抽取
知识图谱
总体国家安全观
South China Sea rights protection archives system
fine-grained South China Sea archival data
information extraction
knowledge graph
a holistic approach to national security
作者简介
彭玉芳(1987-),女,江苏宿迁人,讲师,主要从事安全情报、竞争情报、南海维权证据链、文本挖掘研究;王妍睿(2001-),女,湖北宜昌人,本科生,主要从事信息组织研究;杨海平(1967-),男,江苏大丰人,教授,主要从事知识组织与知识服务、竞争情报与战略管理等研究;徐浩(1989-),男,江苏盐城人,副教授,主要从事智能信息检索;孙宁(2004-),女,黑龙江双鸭山人,本科生,主要从事南海数字资源融合研究。
