摘要
针对现有地理语义相似度计算模型欠考虑实体层次影响因素问题,提出了一种引入实体的地理语义相似度混合计算模型.该模型在已有研究基础上,综合分析了语义相似度概念间的关系、本体结构、属性和实体的几何特征等影响因素,并用各影响因素的综合相似度结果度量地理概念间的语义相似度.最后,通过计算土地利用类型概念语义相似度验证了模型的有效性.
Although some computational models for geo-semantic similarity measurement have been proposed in literature,these models overlook the effect of geospatial entities on similarity.This paper presents a geo-semantic similarity measurement approach that extends the existing approaches by computing geometric feature similarity of geospatial entities.This approach is based on comprehensive analysis of factors on ontology structure,properties of concept and geometric features similarity for geospatial entities.An assessment calculating the semantic similarity of the concept of land use types is carried out to verify the validity of this approach.
出处
《浙江大学学报(理学版)》
CAS
CSCD
北大核心
2018年第2期200-208,共9页
Journal of Zhejiang University(Science Edition)
基金
国家自然科学基金资助项目(41671391
41471313)
测绘地理信息公益性行业科研专项(201512024)
国家科技基础性工作专项(2012FY112300)
关键词
本体
语义相似度
地理实体
土地利用类型
ontology
sematic similarity
geographic entities
land use types