SCIENTIA SINICA Informationis, Volume 50 , Issue 7 : 1019-1032(2020) https://doi.org/10.1360/SSI-2019-0269

Spatiotemporal features based geographical knowledge graph construction

More info
  • ReceivedDec 1, 2019
  • AcceptedJan 20, 2020
  • PublishedJul 6, 2020


Geographical knowledge is the cognitive result of human beings on the spatial distribution, evolution, and interaction of geographical phenomena or things. At present, there is a common phenomenon, i.e., “mass of data, information explosion, and hard to find knowledge” in geographic information services in the context of big data. The geographic knowledge graph (GeoKG) is a knowledge-based system that uses the semantic network to describe geographic concepts, entities, and their relations. It has application prospects in geographic knowledge understanding, geological problem solution, as well as spatiotemporal prediction and decision-making. In addition to the characteristics of general knowledge, geographic knowledge also has specific characteristics of spatiotemporal and geoscience mechanism. Therefore, the construction and applications of GeoKGs have both generality and certain professional specialty. Based on the spatiotemporal characteristics of geographic knowledge and the structure of knowledge graphs, we propose a method for GeoKG construction that considers the characteristics of time and space. Firstly, we describe the basic ideas and technical processes of GeoKG construction and briefly explain the main research contents of geographic knowledge acquisition, geographic knowledge abstraction and representation, as well as geographic knowledge organization and management. Secondly, based on the analysis of basic questions answered by geography, we summarize and abstract geographic knowledge elements and then explore a geographic knowledge representation model with three levels of “geographical concepts-geographic entities-geographical relations" to describe semantic units of geographical knowledge and their logic relations. Finally, the spatiotemporal features are assumed as the basic conditions, and based on the statuses of geographic entities, we propose a formal method based on “process-relation" to represent the evolution process of geographic entities and complex geographic relations. In conclusion, our results can effectively solve the basic problems of GeoKG construction, which are crucial for the acquisition, fusion, reasoning, and application of geographic knowledge. Furthermore, it can be extended in geoscience fields such as geology, environmental science, and meteorology, and has the prospects for the promotion of geoscience knowledge services.

Funded by



[1] James, P. Knowledge graphs. In: Van de Riet, Meersman, eds. Linguistic Instruments in Knowledge Engineering. Amsterdam: Elsevier, 1992. 97-117. Google Scholar

[2] Zhao J, Liu K, He S Z, et al. Knowledge map. Beijing: Higher Education Press, 2018, pp.293. Google Scholar

[3] Chen J, Liu W Z, Wu H, et al. Basic Issues and Research Agenda of Geospatial Knowledge Service. Geomat Inform Sci Wuhan Univ, 2019, 44(01): 38-47. Google Scholar

[4] Chen J Y, Deng S M, Chen H J. CrowdGeoKG: crowdsourced Geo-knowledge graph. In: Proceedings of China Conference on Knowledge Graph and Semantic Computing. Singapore: Springer, 2017. Google Scholar

[5] Ballatore A, Bertolotto M, Wilson D. A Structural-Lexical Measure of Semantic Similarity for Geo-Knowledge Graphs. IJGI, 2015, 4: 471-492 CrossRef ADS Google Scholar

[6] Lu F, Yu L, Qiu P Y. On Geographic Knowledge Graph. J Geo-Inf Sci, 2017, 19(6): 723-734. Google Scholar

[7] Lin H, You L, Hu C B, et al. Prospect of Geo-Knowledge Engineering in the Era of Spatio-Temporal Big Data. Geomat Inform Sci Wuhan Univ, 2018, 43(12): 2205-2211. Google Scholar

[8] Miao Q, Xu P, Li X. The Recognition of the Point Symbols in the Scanned Topographic Maps. IEEE Trans Image Process, 2017, 26: 2751-2766 CrossRef PubMed ADS Google Scholar

[9] Yang D. Research on Map Symbol Recognition Based on Neural Network. Dissertation for MA Degree. Guiyang: Guizhou University, 2007. Google Scholar

[10] Hoermann S, Henzler P, Bach M, et al. Object detection on dynamic occupancy grid maps using deep learning and automatic label generation. In: Li L X, Wang F Y, Loannou P, et al. Proceedings of IEEE Intelligent Vehicles Symposium. Changshu: Institute of Electrical and Electronics Engineers Inc, 2018. 826-833. Google Scholar

[11] Luo X, Zhang J, Cao X. Object-aware power line detection using color and near-infrared images. IEEE Trans Aerosp Electron Syst, 2014, 50: 1374-1389 CrossRef ADS Google Scholar

[12] Tian K. Research on Point Symbols Recognition in Color Topographic Map. Dissertation for MA Degree. Xian: Xidian University, 2017. Google Scholar

[13] Feldman J. Perceptual grouping by selection of a logically minimal model. Int J Comput Vision, 2003, 55(1): 5-25. Google Scholar

[14] Hancer E, Samet R. Advanced contour reconnection in scanned topographic maps. In:Kalagiakos, Panagiotis, Karampelas, et al., eds. Proceedings of International Conference on Application of Information & Communication Technologies, New York:IEEE Computer Society, 2011.1-5. Google Scholar

[15] Pezeshk A, Tutwiler R L. Automatic Feature Extraction and Text Recognition From Scanned Topographic Maps. IEEE Trans Geosci Remote Sens, 2011, 49: 5047-5063 CrossRef ADS Google Scholar

[16] Le X Q, Yang C J, Yu W Y. Spatial Concept Extraction Based on Spatial Semantic Role in Natural Language. Geomat Inform Sci Wuhan Univ, 2005, 30(12): 1100-1103. Google Scholar

[17] Jones C B, Purves R S. Geographical information retrieval. Int J Geographical Inf Sci, 2008, 22: 219-228 CrossRef Google Scholar

[18] Qian J, Zhang Y J, Zhang T. Research on Chinese Person Name and Location Name Recognition Based on Maximum Entropy Model. J Chin Mini-Micro Comput Syst, 2006, 27(09): 1761-1765. Google Scholar

[19] Jiang W M. Research on Extraction of Spatial Orientation Relations for Chinese Texts. Dissertation for Ph.D. Degree. Nanjing: Nanjing Normal University, 2010. Google Scholar

[20] Wu L, Liu L, Li H R, et al. A Chinese Toponym Recognition Method Based on Conditional Random Field. Geomat Inform Sci Wuhan Univ, 2017, 42(02): 150-156. Google Scholar

[21] Feng Y Y, Sun L, Zhang D K, et al. Study on Chinese Named Entity Recognition Using Small Scale Character Tail Hints. Acta Electron Sin, 2008 (09): 1833-1838. Google Scholar

[22] Zhang C J. Study on the analysis method of event space-time and attribute information in Chinese text. Dissertation for Ph.D. Degree. Nanjing: Nanjing Normal University, 2013. Google Scholar

[23] Tang X R, Chen X H, Zhang X Y. Research on Toponym Resolution in Chinese Text. Geomat Inform Sci Wuhan Univ, 2010, 35(08): 930-935. Google Scholar

[24] Zhang C J, Zhang X Y, Chen Y T, et al. Extraction of geographical attribute-values in natural language. In: Lee G, eds. Advances in Computational Environment Science. Melbourne: Springer Verlag, 2012.51-59. Google Scholar

[25] Zhang X Y, Ye P, Wang S, et al. Geological entity recognition method based on Deep Belief Networks. Acta Petrol Sin. 2018, 34(2): 343-351. Google Scholar

[26] Wang S, Zhang X Y, Ye P, et al. Deep belief networks based toponym recognition for Chinese text. Isprs Int J Geo-Inf, 2018, 7. Google Scholar

[27] Zhang X Y. Spatio-temporal Information Extraction in Chinese Text. Commun China Comput Feder, 2015, 11(11):33-39. Google Scholar

[28] Zhang C J, Zhang X Y, Li M, et al. Interpretation of Temporal Information in Chinese text. Geogr Geo-Inf Sci, 2014, 30(6): 1-7. Google Scholar

[29] Ma L L, Li H C, Wei Y, et al. Chinese Text Temporal Expression Recognition and Normalization Method Based on Rules. J Inf Eng Univ, 2017, 18(5): 560-565. Google Scholar

[30] Yan Z F, Ji D H. Exploraton of Chinese temporal information extraction based on CRF and semi-supervised learning. Comput Eng Desig, 2015,36(6): 1642-164. Google Scholar

[31] Liu Z J, Tang B Z, Wang X L, et al. CMedTEX: A rule-based temporal expression extraction and normalization system for Chinese clinical notes. Annu Symp Proc, 2017: 818-826. Google Scholar

[32] Jin B W. Study on Chinese Temporal Expression Recognition and Normalization. Dissertation for MA Degree. Shenyang: Shenyang University of Aeronautics and Astronautics, 2018. Google Scholar

[33] Hao T, Pan X, Gu Z. A pattern learning-based method for temporal expression extraction and normalization from multi-lingual heterogeneous clinical texts.. BMC Med Inform Decis Mak, 2018, 18: 22 CrossRef PubMed Google Scholar

[34] Montello D R, Goodchild M F., Gottsegen J, et al. Where's downtown? Behavioral Methods for determining referents of vague spatial queries. Spat Cogn Comput, 2003, 3: 185-204. Google Scholar

[35] Liu Y, Yuan Y H, Zhang Y.A Cognitive Approach to Modeling Vague Geographical Features: A Case Study of Zhongguancun. J Remote Sens, 2008, 2: 370-377. Google Scholar

[36] Yan F. Estimation of Spatial Range of Fuzzy Location Based on Interest Points. Dissertation for MA Degree. Shenyang: Shenyang University of Aeronautics and Astronautics, 2018. Google Scholar

[37] Vogele T, Schlieder C, Visser U. Intuitive modelling of place name regions for spatial information retrieval. In: Kuhn W, Michael W, Timpf S, eds. Spatial Information Theory Foundations of Geographic Information Science. Germany: Springer-Verlag, 2003. 239-252. Google Scholar

[38] Zhang X Y, Zhu S N. Contextual spatial relations based spatial modelling of vague place names. In: Luo Q, eds. 2010 2nd IITA International Conference on Geoscience and Remote Sensing. New York: IEEE Computer Society, 2010. 23-26. Google Scholar

[39] Liu S Y. Extracting Landslide Disaster Information from Web Pages. Dissertation for MA Degree. Chengdu: Southwest Jiaotong University, 2015. Google Scholar

[40] Imran M, Elbassuoni S, Castillo C, et al. Extracting information nuggets from disaster-related messages in social media. In:Comes T, Fiedrich F, Fortier S and et al.,eds. 10th International Conference on Information Systems for Crisis response and management. Germany:KIT,2013.791-801. Google Scholar

[41] Yu L, Lu F, Zhang H C. Extracting Geographic Information from Web Texts:Status and Development. J Geo-Inf Sci, 2015,17(02): 127-134. Google Scholar

[42] Matsuo Y, Sakaki T, Uchiyama K, et al. Graph-based word clustering using a web search engine. In: Jurafsky D, Gaussier E, eds. Proceedings of the 2006 Conference on Empirical Methods in Natural Language Processing. Sydney: ACT, 2006.542-550. Google Scholar

[43] Su F L, Xie Q H, Qiu J Y, et al. Study on word clustering for attribute extraction based on deep learning. Micro comput ITS Appl, 2016,35 (01): 53-55 + 59. Google Scholar

[44] Jiang H J. Slot Filling Via Deep Learning. Dissertation for MA Degree. Zhejiang: Zhejiang University, 2017. Google Scholar

[45] Allen J F. Towards a general theory of action and time. Artificial Intelligence, 1984, 23: 123-154 CrossRef Google Scholar

[46] 朱少楠, 张雪英, 张春菊. 地理空间关系描述的句法模式识别. In:Tan H H,eds.Proceedings of 2010 International Conference on Broadcast Technology and Multimedia Communication(Volume 4). Hong Kong: International Communication Sciences Association, 2010. 367-370. Google Scholar

[47] Ma L B, Gong J Y. Research on Spatial Database Query Oriented Natural Language. Comput Eng Appl, 2003 (22): 16-19. Google Scholar

[48] Zheng Y, Long Y, Ming X N, et al. Natural Language Description of Geographic Location Considering Various Spatial Relations with Different References. J Geo-Inf Sci, 2011,13 (04): 465-471. Google Scholar

[49] Yu L, Lu F, Liu X L. A Bootstrapping Based Approach for Open Geo-entity Relation Extraction. Acta Geod Cartogr Sin, 2016, 45(05): 616-622. Google Scholar

[50] Zhang X Y, Zhu S N, Zhang C J. Annotation of Geographical Named Entities of Chinese Text. Acta Geod Cartogr Sin, 2012a, 41 (01): 115-120. Google Scholar

[51] Zhang X Y, Zhang C J, Zhu S N. Annotation of Geographical Named Entities of Chinese Text. Acta Geod Cartogr Sin, 2012b, 41(03): 468-474. Google Scholar

[52] Zhang X Y, Zhang C J, Du C L, et al. SVM based extraction of spatial relations in text. In:Leung L, Zhou C H, Lees B, et al.,eds.Proceedings of Spatial Data Mining and Geographical Knowledge Services. Beijing: Institute of Electrical and Electronics Engineers 2011.529-533. Google Scholar

[53] Wang J B, Lu F, Wu S, et al. Constructing the corpus of Geographical entity relations based on automatic annotation. J Geo-inf Sci, 2018,20(07): 871-879. Google Scholar

[54] Jin Y. Research and implementation of geographical relationship extraction system based on deep learning. Dissertation for MA Degree. Beijing: Beijing University of Posts and Telecommunications,2019. Google Scholar

[55] Gao J L, Yu L, Qiu P Y, et al. A knowledge-based method for filtering Geo-entity relations. J Geo-Inf Sci, 2019, 21(9): 1392-1401. Google Scholar

[56] Brodt A, Nicklas D, Mitschang B. Deep integration of spatial query processing into native RDF triple stores. In: Agrawal D, Zhang P S, Abbadi A E, et al., eds. Proceedings of the 18th SIGSPATIAL International Conference on Advances in Geographic Information Systems, California: ACM, 2010, 33-42. Google Scholar

[57] Liagouris J, Mamoulis N, Bouros P. An effective encoding scheme for spatial RDF data. Proc VLDB Endow, 2014, 7: 1271-1282 CrossRef Google Scholar

[58] Wang D, Zou L, Feng Y, et al. S-store: An engine for large RDF graph integrating spatial information. In: Meng W Y, Feng L, Bressan S, et al., eds. Proceedings of the 18th International Conference on Database Systems for Advanced Applications (DASFAA 2013), Berlin: Springer Verlag, 2013, 31-47. Google Scholar

[59] Fu B J. Geography: From Knowledge, Science to decision making support. Acta Geogr Sin, 2017,72(11): 1923-1932. Google Scholar

[60] Commission on Geographical Education, International Geography Union. International Charter on Geography Education. 1992. Google Scholar

[61] Claramunt C, Parent C, Theriault M. Design patterns for spatio-temporal processes.In:Mitchell R, Jezequel J M, Bosch J, et al., eds. Proceedings Technology of Object-Oriented Languages and Systems. California: IEEE, 1999. 2-15. Google Scholar

[62] Haarslev V. A description logic with concrete domains and a role-forming predicate operator. J Logic Computation, 1999, 9: 351-384 CrossRef Google Scholar

[63] Wang S. Research on Construction of Geographic Knowledge Graph Driven by Natural Language. Dissertation for Ph.D. Degree. Nanjing: Nanjing Normal University, 2018. Google Scholar

[64] Frank, Andrew U. Qualitative spatial reasoning: cardinal directions as an example. Int J Geogr Inform Syst, 1996, 10(3): 269-290. Google Scholar

[65] Kondaveeti A, Runger G C, Liu H, et al. Extracting geographic knowledge from sensor intervention data using spatial association rules. In: Proceedings of IEEE International Conference on Spatial Data Mining & Geographical Knowledge Services. Beijing: Institute of Electrical and Electronics Engineers, 2011.127-130. Google Scholar

[66] Yu C, Peuquet D J. A GeoAgent?based framework for knowledge?øriented representation: Embracing social rules in GIS. Int J Geographical Inf Sci, 2009, 23: 923-960 CrossRef Google Scholar

[67] Otero-Cerdeira L, Rodríguez-Martínez F J, Gómez-Rodríguez A. Ontology matching: A literature review. Expert Syst Appl, 2015, 42: 949-971 CrossRef Google Scholar

[68] Jiang B C, Wan G, Xu J, et al. Geographic Knowledge Graph Building Extracted Multi-sourced Heterogeneous Data. Acta Geod Cartogr Sin, 2018, 47(8): 1051-1061. Google Scholar

Copyright 2020  CHINA SCIENCE PUBLISHING & MEDIA LTD.  中国科技出版传媒股份有限公司  版权所有

京ICP备14028887号-23       京公网安备11010102003388号