基于地理位置信息的知识图谱查询方法

doi:10.16180/j.cnki.issn1007-7820.2022.12.003

摘要/Abstract

摘要：

现有知识图谱查询方法忽略了实体本身的地理位置信息,故而不支持地理位置相关查询的问题。针对该问题,在融合地理位置信息的混合知识图谱的基础上,文中提出一种基于地理位置信息的知识图谱查询方法。通过抽取查询问题中的三元组,构建了相应的查询图以理解自然语言查询问题。将基于地理位置信息的查询问题分为6类,再结合已有事实类问题的语义查询方法,分别根据查询图或K近邻搜索思想研究相应的知识图谱查询方法。实验结果表明,文中所提方法的精确率可达77%以上,能够为基于地理位置信息的查询提供有效支持。

关键词: 知识图谱, 地理位置信息, 地理实体, 百科知识, 图查询, 语义查询, 三元组抽取, 查询图

Abstract:

Existing knowledge graph query methods ignore the geographic location information of entities themselves, so they do not support geographic location related queries. In view of this problem, on the basis of the hybrid knowledge graph integrating geographic location information, this study proposes a knowledge graph query method based on geographic location information. By extracting the triples from the query problem, the corresponding query graph is constructed to understand the natural language query problem. The query problems based on geographic location information are divided into six categories, and combined with the existing semantic query methods for fact-based problems, the corresponding knowledge graph query methods are studied according to the query graph or K-nearest neighbor search idea. Experimental results show that the accuracy rate of the proposed method can reach more than 77%, which can provide effective support for query based on geographic location information.

Key words: knowledge graph, geographic location information, geographic entity, encyclopedic knowledge, graph query, semantic query, triple extraction, query graph

中图分类号:

TP399

李怡霈,王宇翔. 基于地理位置信息的知识图谱查询方法[J]. 电子科技, 2022, 35(12): 17-25.

LI Yipei,WANG Yuxiang. Knowledge Graph Query Method Based on Geographic Location Information[J]. Electronic Science and Technology, 2022, 35(12): 17-25.

图/表 10

图1

表1

表2

表3

图2

图3

图4

图5

图6

图7

参考文献 19

[1]	王晓阳, 郑骁庆, 肖仰华. 智慧搜索中的实体与关联关系建模与挖掘[J]. 通信学报, 2015, 36(12):17-27.
	Wang Xiaoyang, Zheng Xiaoqing, Xiao Yanghua. Entity-relation modeling and discovery for smart search[J]. Journal of Communications, 2015, 36(12):17-27.
[2]	Peng P, Zou L, Guan R. Accelerating partial evaluation in distributed SPARQL query evaluation[C]. Macau: Proceedings of the IEEE Thirty-fifth International Conference on Data Engineering, 2019.
[3]	Han S, Zou L, Yu J X, et al. Keyword search on RDF graphs-a query graph assembly approach[C]. Singapore: Proceedings of the ACM on Conference on Information and Knowledge Management, 2017.
[4]	Zou L, Huang R, Wang H, et al. Natural language question answering over RDF: A graph data driven approach[C]. Snowbird: Proceedings of the ACM International Conference on Management of Data, 2014.
[5]	Zheng W, Yu J X, Zou L, et al. Question answering over knowledge graphs: Question understanding via template decomposition[J]. Proceedings of the Very Large Data Base Endowment, 2018, 11(11):1373-1386.
[6]	Huang X, Zhang J, Li D, et al. Knowledge graph embedding based question answering[C]. Melbourne: Proceedings of the Twelfth ACM International Conference on Web Search and Data Mining, 2019.
[7]	Khan A, Wu Y, Aggarwal C C, et al. NeMa:Fast graph search with label similarity[J]. Proceedings of the Very Large Data Base Endowment, 2013, 6(3):181-192.
[8]	Jin J, Khemmarat S, Gao L, et al. Querying web-scale information networks through bounding matching scores[C]. Firenze: Proceedings of the Twenty-fourth International Conference on World Wide Web, 2015.
[9]	Gao J, Liu P, Kang X, et al. PRS: Parallel relaxation simulation for massive graphs[J]. The Computer Journal, 2016, 59(6):848-860. doi: 10.1093/comjnl/bxu159
[10]	Hong L, Zou L, Lian X, et al. Subgraph matching with set similarity in a large graph database[J]. IEEE Transactions on Knowledge and Data Engineering, 2015, 27(9):2507-2521. doi: 10.1109/TKDE.2015.2391125
[11]	Mottin D, Lissandrini M, Velegrakis Y, et al. Exemplar queries: a new way of searching[J]. The Very Large Data Base Journal, 2016, 25(6):741-765.
[12]	Zheng W, Zou L, Lian X, et al. Graph similarity search with edit distance constraint in large graph databases[C]. San Francisco: Proceedings of the Twenty-second ACM International Conference on Information and Knowledge Management, 2013.
[13]	Zheng W, Zou L, Peng W, et al. Semantic SPARQL similarity search over RDF knowledge graphs[J]. Proceedings of the Very Large Data Base Endowment Endowment, 2016, 9(11):840-851.
[14]	Shekarpour S, Marx E, Auer S, et al. RQUERY: Rewriting text queries to alleviate the vocabulary mismatch problem on RDF knowledge bases[C]. San Francisco: Proceedings of the Thirty-first AAAI Conference on Artificial Intelligence, 2017.
[15]	Auer S, Lehmann J, Hellmann S. Linkedgeodata:Adding a spatial dimension to the web of data[C]. Heidelberg: Proceedings of the International Semantic Web Conference, 2009.
[16]	Stadler C, Lehmann J, K Höffner, et al. Linkedgeodata: A core for a web of spatial open data[J]. Semantic Web, 2012, 3(4):333-354. doi: 10.3233/SW-2011-0052
[17]	Chen J, Deng S, Chen H. Crowdgeokg:Crowdsourced geo-knowledge graph[C]. Singapore: Proceedings of the China Conference on Knowledge Graph and Semantic Computing, 2017.
[18]	Wang Y X, Khan A, Wu T X, et al. Semantic guided and response times bounded top-k similarity search over knowledge graphs[C]. Dallsa: Proceedings of the IEEE Thirty-sixth International Conference on Date Engineering, 2020.
[19]	Bordes A, Usunier N, Garcia-Duran A, et al. Translating embeddings for modeling multi-relational data[C]. Lake Tahoe: Neural Information Processing Systems, 2013.

位置关系	中文位置关系	备注
nextTo	邻近	距离
nearBy	附近	距离
notFarFrom	不远	距离
notAround	稍远	距离
littleFarFrom	较远	距离
northOf	向南	方位
southOf	向北	方位
westOf	向东	方位
eastOf	向西	方位
northwestOf	东南	方位
southwestOf	东北	方位
northeastOf	西南	方位
southeastOf	西北	方位

查询问题	初始抽取结果	校正后抽取结果
Q₁	(嘉里中心,北边的,医院)(医院,上班,牙科医生)	(嘉里中心,北边的,医院)(医院,牙科医生,?)
Q₂	(春波南苑,附近的,装修公司)(装修公司,工作,水电师傅)	(春波南苑,附近的,装修公司)(装修公司,水电师傅,?)
Q₃	(海创基地,1 000,范围)(沈思齐,的,校友)	(海创基地,1 000米范围内,住宅区)(沈思齐,校友,?)
Q₄	(杭州大剧院,步行,15,学校)(邵楠,的,熟人)	(杭州大剧院,步行15分钟内,学校)(邵楠,熟人,?)
Q₅	(太子湾公园,最近,公司)	(太子湾公园,最近的,公司)
	(吕乐,的,同学)	(吕乐,同学,?)
Q₆	(嘉柯宠物诊所,隔壁的,打印店)(打印店,的,营业时间)	(嘉柯宠物诊所,隔壁的,打印店)(打印店,营业时间,?)

问题分类	查询问题示例	序号	备注
基于地理位置信息的实体查询	在嘉里中心北边的医院上班的牙科医生	Q₁	基于方位关系
	在春波南苑附近的装修公司工作的水电师傅	Q₂	基于距离关系
	住在海创基地1 000米范围内的沈思齐的校友	Q₃	直接范围约束
	步行15分钟内能从学校到杭州大剧院的邵楠的熟人	Q₄	间接范围约束
	公司距离太子湾公园最近的吕乐寅的同学	Q₅	最值范围约束
基于地理位置信息的属性查询	嘉柯宠物诊所隔壁的打印店的营业时间	Q₆	基于位置关系