Education, Science, Technology, Innovation and Life
Open Access
Sign In

Geographic Information Linking Method Based on Machine Learning

Download as PDF

DOI: 10.23977/infkm.2022.030204 | Downloads: 8 | Views: 456

Author(s)

Jun Shu 1

Affiliation(s)

1 Wuhan University of Engineering Science, Wuhan, Hubei, 430200, China

Corresponding Author

Jun Shu

ABSTRACT

With the continuous development of Internet geographic information(GI) system, various GI sources have sprung up. However, the representation and storage methods of GI of these GI data sources are different, resulting in significant differences in the accuracy and integrity of GI, which has caused many difficulties for GI integration. Therefore, this paper introduces the importance of geospatial relationship in GI data link, and discusses the GI link method based on machine learning(ML); The link results of support vector machine classification method and k-nearest neighbor classification method in GI data link method are discussed respectively. Both of them have achieved good experimental results, and the differences between them are analyzed in detail. At the same time, the GI data link method in this paper is compared with the geographic context related information link method using graph theory method. The experimental results show that the GI data link method proposed in this paper is obviously better than this method, which further proves the accuracy and effectiveness of the GI data link method proposed in this paper. 

KEYWORDS

Machine Learning, Geographic Information, Geographic Features, Linking Methods

CITE THIS PAPER

Jun Shu, Geographic Information Linking Method Based on Machine Learning. Information and Knowledge Management (2022) Vol. 3: 23-29. DOI: http://dx.doi.org/10.23977/infkm.2022.030204.

REFERENCES

[1] Tempelmeier N, Demidova E. Linking Open Street Map with knowledge graphs — Link discovery for schema-agnostic volunteered GI[J]. Future Generation Computer Systems, 2021, 116(2):349-364.
[2] Kaya L A, E Kut Grgün. Land use and land cover change monitoring in Bandrma (Turkey) using remote sensing and GI systems [J]. Environmental Monitoring and Assessment, 2020, 192(7):1-18.
[3] Bayazidy-Hasanabad M, Vayghan S S, Ghasemkhani N, et al. Developing a volunteered GI-based system for rapidly estimating damage from natural disasters[J]. Arabian Journal of Geosciences, 2021, 14(17):1-13.
[4] Moreri K K. Using Kappa methodology to consider volunteered GI in official land administration systems in developing countries [J]. Spatial Information Research, 2020, 28(3):299-311.
[5] Saqr A M, Ibrahim M G, Fujii M, et al. Sustainable Development Goals (SDGs) Associated with Groundwater Over-Exploitation Vulnerability: GI System-Based Multi-criteria Decision Analysis[J]. Natural Resources Research, 2021, 30(6):4255-4276.
[6] Dehghani M H, Asghari F B, Mahvi A H, et al. Spatiotemporal variation of drying and salinity water basin on the quality of coastal aquifers using GI system[J]. Environmental Geology, 2019, 78(15):444.1-444.13.
[7] Alberto R T, Biagtan A R, Isip M F, et al. Hot spot area analysis of onion armyworm outbreak in Nueva Ecija using GI system[J]. Spatial Information Research, 2019, 27(6):673-680.
[8] Owolabi S T, K Madi, Kalumba A M, et al. A groundwater potential zone mapping approach for semi-arid environments using remote sensing (RS), GI system (GIS), and analytical hierarchical process (AHP) techniques: a case study of Buffalo catchment, Eastern Cape, South Africa[J]. Arabian Journal of Geosciences, 2020, 13(22):1-17.
[9] Sk M M, Ali S A, Ahmad A. Optimal Sanitary Landfill Site Selection for Solid Waste Disposal in Durgapur City Using GI System and Multi-criteria Evaluation Technique[J]. KN - Journal of Cartography and GI, 2020, 70(4):163-180.
[10] Peethambaran B, Anbalagan R, Shihabudheen K V, et al. Robustness evaluation of fuzzy expert system and extreme learning machine for GI system-based landslide susceptibility zonation: A case study from Indian Himalaya[J]. Environmental Geology, 2019, 78(6):231.1-231.20.
[11] Comber S, D Arribas‐Bel. ML innovations in address matching: A practical comparison of word2vec and CRFs[J]. Transactions in GIS, 2019, 23(2):334-348.
[12] Borodinov A A. Development and research of algorithms for determining user preferred public transport stops in a GI system based on ML methods[J]. Computer Optics, 2020, 44(4):646-652.

All published work is licensed under a Creative Commons Attribution 4.0 International License.

Copyright © 2016 - 2031 Clausius Scientific Press Inc. All Rights Reserved.