This paper focuses on the global LNG trade network from 2013 to 2022, aiming to identify potential LNG trade relationships through an improved link prediction algorithm. It is important to assist LNG-consuming countries in finding potential partners and ensuring energy supply security. External factors, such as geographical distance between nodes, trade volumes and poli-tical stability of countries, are fully considered, and these are coupled with internal network indices through a multi-index approach. A random forest classifier is used to enhance prediction accuracy. The results show that the average prediction accuracy can reach 97.55%. And, 120 potential links had been successfully identified, of which 99 was transformed into real links in 2022, and 94 links were transformed into real links within following three years. These results not only confirm the high precision of the algorithm in forecasting unknown links but also highlight its effective capacity for identifying potential future links.
YU Hongchu
,
CHEN Feng
,
ZHANG Ping
. Link Prediction Method of LNG Shipping Network Based on Multi-index Coupling Learning[J]. Complex Systems and Complexity Science, 2026
, 23(4)
: 9
-18
.
DOI: 10.13306/j.1672-3813.2026.04.002
[1] Mafakheri A, Sulaimany S, Mohammadi S. Predicting the establishment and removal of global trade relations for import and export of petrochemical products[J]. Energy, 2023, 269: 126850.
[2] Filimonova I V, Komarova A V, Sharma R, et al. Transformation of international liquefied natural gas markets:new trade routes[J]. Energy Reports, 2022, 8: 675-682.
[3] Guan Q, An H, Gao X, et al. Estimating potential trade links in the international crude oil trade: a link prediction approach[J]. Energy, 2016, 102: 406-415.
[4] Feng S, Li H, Qi Y, et al. Who will build new trade relations? finding potential relations in international liquefied natural gas trade[J]. Energy, 2017, 141: 1226-1238.
[5] Zhou X, Zhang H, Zheng S, et al. The crude oil international trade competition networks:evolution trends and estimating potential competition links[J]. Energies, 2022, 15(7): 2395.
[6] Dong Z. Who will trade bauxite with whom? finding potential links through link prediction[J]. Resources Policy, 2019, 63: 101417.
[7] Liu S, Dong Z, Ding C, et al. Do you need cobalt ore? estimating potential trade relations through link prediction[J]. Resources Policy, 2020, 66: 101632.
[8] Zhu M, Zhou X, Zhang H, et al. International trade evolution and competition prediction of boron ore: based on complex network and link prediction[J]. Resources Policy, 2023, 82: 103542.
[9] Jin Y, Yang Y, Liu W. Findingglobal liquefied natural gas potential trade relations based on improved link prediction[J]. Sustainability, 2022, 14(19): 12403.
[10] 陈广福,王海波.基于聚类信息和对称非负矩阵分解的链路预测模型研究[J].计算机应用研究,2021,38(12):3733-3738.
Chen G F, Wang H B. Link prediction model based on clustering information and symmetric non-negative matrix factorization[J]. Application Research of Computers, 2021,38(12):3733-3738.
[11] Anand S, Rahul, Mallik A, et al. Integrating node centralities, similarity measures, and machine learning classifiers for link prediction[J]. Multimedia Tools and Applications, 2022, 81(27): 38593-38621.
[12] 闫兆进,杨慧.基于多源数据和船舶停留轨迹语义建模的港口目标识别[J].地球信息科学学报,2022,24(9):1662-1675.
Yan Z J, Yang H. Harbor detection based on multi-source data and semantic modeling of ship stop trajectory[J]. Journal of Geo-information Science, 2022,24(9):1662-1675.
[13] 吕琳媛. 复杂网络链路预测[J]. 电子科技大学学报, 2010, 39(5): 651-661.
Lü L Y. Link prediction on complex networks[J]. Journal of University of Electronic Science and Technology of China, 2010, 39(5): 651-661.
[14] Liu W P, Lyu L Y. Link prediction based on local random walk[J]. Europhysics Letters, 2010, 89(5): 58007.
[15] 吕亚楠,韩华,贾承丰,等. 基于有偏向的重启随机游走链路预测算法[J]. 复杂系统与复杂性科学, 2018, 15(4): 17-24.
Lü Y N, Han H, Jia C F, et al. Link prediction algorithm based on biased random walk with restart[J]. Complex Systems and Complexity Science, 2018, 15(4): 17-24.
[16] Barnes R, Bosworth R. LNG is linking regional natural gas markets: evidence from the gravity model[J]. Energy Economics, 2015, 47: 11-17.
[17] Magnier H J, Jrad A. A minimal simplified model for assessing and devising global LNG equilibrium trade portfolios while maximizing energy security[J]. Energy, 2019, 173: 1221-1233.
[18] 梅强, 胡勤友, 刘希亮, 等. 全球LNG海上运输网络演化及中国贸易现状分析[J]. 地球信息科学学报, 2022, 24(9):1701-1716.
Mei Q, Hu Q Y, Liu X L, et al. Research on the evolution of Global LNG maritime transportation network and trade condition of China[J]. Journal of Geo-information Science, 2022,24(9):1701-1716.
