摘要
针对通用检索系统应用于灾害应急领域时,存在低效、复杂、不专业等问题,该文以火灾应急专题图组为例,构建了相关本体知识库,对火灾发生前后所需的专题图进行了多角度、全场景的系统归纳,并针对该知识库提出了一种基于本体的扩散检索方法,用以满足用户的检索需求。该方法采用word2vec模型为相似度计算工具,通过训练相关的火灾应急语料库,提高了计算的精度。此外,根据本体的结构特点,采用分段激活扩散算法,实现了基于关键字的语义扩散检索。通过程序的验证结果表明,该方法不仅可以根据用户检索需求快速有效地在海量图组中检索出所需专题图信息,还可以发掘出用户潜在兴趣点,从而提高检索精度。
In view of the problem that the inefficient, complex and unprofessional issues arise when the general retrieval system is applied to the field of disaster emergency response, by taking fire emergency thematic map as an example, this paper constructed the relevant ontology knowledge base, made a multi-angle and whole-scene systematic summary of the thematic maps needed before and after the fire, and put forward a ontology-based spreading retrieval method to meet users' retrieval needs. In the method, the word2vec model was used as the similarity calculation tool, and the accuracy of the calculation was improved by training the relevant fire emergency corpus. In addition, according to the structural characteristics of the ontol-ogy, the segmented spreading activation algorithm was used to implement semantic spreading retrieval based on keywords. The results of the program showed that the method could not only quickly and effectively retrieve the required thematic map information in the mass map group according to the user's retrieval requirements, but also discover the potential interest points of the user, thereby improving the retrieval accuracy.
作者
冯天文
李轶鲲
刘涛
杜萍
杨国林
FENG Tianwen;LI Yikun;LIU Tao;DU Ping;YANG Gaolin(Faculty of Geomatics,Lanzhou Jiaotong University.Lanzhou 730070,China;Gansu Provincial Engineering Laboratory for National Geographic State Monitoring,Lanzhou 730070,China)
出处
《测绘科学》
CSCD
北大核心
2018年第12期111-117,共7页
Science of Surveying and Mapping
基金
国家重点研发计划课题项目(2016YFC0803106)
国家自然科学基金项目(41761088,41764001)
兰州交通大学优秀平台支持项目(201806)
关键词
灾害应急
相似度
语义扩散
海量图组
disaster emergency
similarity
semantic spreading
mass graph group
