计算机科学

计算机科学 ›› 2026, Vol. 53 ›› Issue (7): 308-314.doi: 10.11896/jsjkx.250500009

• 计算机网络 • 上一篇    下一篇

基于图神经网络和共形分位数回归的可信IP定位方法

台文鑫1, 刘学婷1, 王枭涵1, 钟婷1, 王永2, 周帆1   

  1. 1 电子科技大学信息与软件工程学院 成都 610054
    2 郑州埃文科技有限公司 郑州 450047
  • 收稿日期:2025-05-06 修回日期:2025-09-10 出版日期:2026-07-15 发布日期:2026-07-10
  • 通讯作者: 周帆(fan.zhou@uestc.edu.cn)
  • 作者简介:(wxtai@std.uestc.edu.cn)
  • 基金资助:
    国家自然科学基金(62176043,62572097)

Trustworthy IP Geolocation Method via Graph Neural Networks and Conformalized Quantile Regression

TAI Wenxin1, LIU Xueting1, WANG Xiaohan1, ZHONG Ting1, WANG Yong2, ZHOU Fan1   

  1. 1 School of Information and Software Engineering,University of Electronic Science and Technology of China,Chengdu 610054,China
    2 Zhengzhou Aiwen Tech Co.,Ltd.,Zhengzhou 450047,China
  • Received:2025-05-06 Revised:2025-09-10 Published:2026-07-15 Online:2026-07-10
  • About author:TAI Wenxin,born in 1997,Ph.D candidate,is a member of CCF(No. A03226G).His main research interests include controllable generative models and trustworthy artificial intelligence.
    ZHOU Fan,born in 1981,Ph.D,professor,Ph.D supervisor.His main research interests include spatial-temporal data analysis,graph learning and network security.
  • Supported by:
    National Natural Science Foundation of China(62176043,62572097).

PDF (PC)

15

摘要： IP定位作为网络空间测绘与管理的核心技术,在网络安全、内容推荐和金融风控等多个领域具有重要应用价值。近年来,神经网络已成为IP定位领域的主流建模范式,相关研究普遍将最小化平均定位误差作为优化目标。然而,在风险敏感场景中,定位算法的误差可控性同样至关重要,单一追求平均误差最小化的建模思路难以满足实际需求。为此,提出一种融合图神经网络与共形分位数回归的可信IP定位方法。该方法区别于传统的点估计范式,能够输出具备置信水平保障的预测区间,从而实现对定位误差范围的可验证控制。在多个真实数据集上的实验结果表明,所提方法在90%置信水平下的预测区间覆盖率偏差小于1%,同时能够保持较窄的区间宽度,有效提升了IP定位算法在风险敏感场景下的可靠性与实用性1)。

关键词: IP定位, 图神经网络, 可信, 共形预测, 分位数回归

Abstract: IP geolocation,as a fundamental technology for cyberspace mapping and management,has significant applications in various domains such as network security,content recommendation,and financial risk control.With the rapid development of artificial intelligence,neural networks have become the predominant modeling paradigm for IP geolocation in recent years,with most existing studies focusing on minimizing the average geolocation error.However,in risk-sensitive scenarios,the controllability of localization errors is equally critical,and modeling paradigms that solely focus on minimizing average errors often fall short of practical requirements.To address this issue,this paper proposes a trustworthy IP geolocation method that integrates graph neural networks with conformalized quantile regression.Unlike traditional point estimation methods,the proposed method produces prediction intervals with guaranteed confidence levels,enabling verifiable control over the range of geolocation errors.Experimental results on multiple real-world datasets demonstrate that the proposed method achieves a prediction interval coverage deviation of less than 1% at the 90% confidence level,while maintaining narrow interval widths,significantly enhancing the trustworthiness and practicality of IP geolocation in risk-sensitive scenarios.

Key words: IP geolocation, Graph neural networks, Trustworthy, Conformal prediction, Quantile regression

中图分类号: 

  • TP391
[1]PADMANABHAN V N,SUBRAMANIAN L.An investigation of geographic mapping techniques for Internet hosts[C]//Proceedings of the 2001 Conference on Applications,Technologies,Architectures,and Protocols for Computer Communications.2001:173-185.
[2]LIU X T,TAI W X,ZHOU F,et al.Survey of IP geolocation:theory,methods,and innovative applications [J].Journal on Communications,2025,46(4):33-48.
[3]LIN J L,LI C L,SONG G L,et al.A survey of client-indepen-dent IP geolocation techniques [J].Journal of Software,2025,36(1):321-340.
[4]JIANG H,LIU Y,MATTEWS J N.IP geolocation estimationusing neural networks with stable landmarks[C]//2016 IEEE Conference on Computer Communications Workshops.IEEE,2016:170-175.
[5]DING S,LUO X,WANG J,et al.GNN-Geo:a graph neural network-based fine-grained IP geolocation framework[J].IEEE Transactions on Network Science and Engineering,2023,10(6):3543-3560.
[6]WANG Z Y,ZHOU F,ZENG W X,et al.Connecting the hosts:Street-level IP geolocation with graph neural networks[C]//Proceedings of the 28th ACM SIGKDD Conference on Know-ledge Discovery and Data Mining.2022:4121-4131.
[7]MA Z R,HU X H,ZHANG S C,et al.SubvectorS_Geo:A neural-network-based IPv6 geolocation algorithm[J].Applied Sciences,2023,13(2):754-772.
[8]MA Z R,HU X H,LI N,et al.HGL_GEO:Finer-grained IPv6 geolocation algorithm based on hypergraph learning[J].Information Processing & Management,2023,60(6):1-17.
[9]TAI W,CHEN B,ZHONG T,et al.Ripgeo:Robust street-level IP geolocation[C]//2023 24th IEEE International Conference on Mobile Data Management.IEEE,2023:138-147.
[10]TAI W X,CHEN B,ZHOU F,et al.Trustgeo:Uncertainty-aware dynamic graph learning for trustworthy ip geolocation[C]//Proceedings of the 29th ACM SIGKDD Conference on Knowledge Discovery and Data Mining.2023:4862-4871.
[11]YANG K,TAI W,LI Z,et al.Exploring self-explainable street-level IP geolocation with graph information bottleneck[C]//ICASSP 2024-2024 IEEE International Conference on Acoustics,Speech and Signal Processing.IEEE,2024:7270-7274.
[12]LIU C,CHENG R,YUAN F,et al.GDD-Geo:IPv6 geolocation by graph dual decomposition[J].Computer Communications,2025,231:108019.
[13]CHENG R,DING S,ZHANG L,et al.IPv6Landmarker:Enhancing IPv6 Street-Level Geolocation through Network Landmark Mining and Targeted Updates[J].IEEE Transactions on Network Science and Engineering,2025,12(2):1280-1269.
[14]GAL Y,GHAHRAMANI Z.Dropout as a Bayesian approximation:Representing model uncertainty in deep learning[C]//Proceedings of the 33rd International Conference on Machine Learning.PMLR,2016:1050-1059.
[15]LAKSHMINARAYANAN B,PRITZEL A,BLUNDELL C.Simple and scalable predictive uncertainty estimation using deep ensembles[C]//Advances in Neural Information Processing Systems(NeurIPS).2017.
[16]WHOIS-Web.com [EB/OL].[2025-04-25].https://www.web.com/whois/index.jsp.
[17]LIU H,ZHANG Y,ZHOU Y,et al.Mining checkins from location-sharing services for client-independent ip geolocation[C]//IEEE INFOCOM 2014-IEEE Conference on Computer Communications.IEEE,2014:619-627.
[18]DAN O,PARIKH V,DAVISON B D.Improving IP geolocation using query logs[C]//Proceedings of the Ninth ACM International Conference on Web Search and Data Mining.2016:347-356.
[19]GUEYE B,ZIVIANI A,CROVELL M,et al.Constraint-based geolocation of internet hosts[C]//Proceedings of the 4th ACM SIGCOMM Conference on Internet Measurement.2004:288-293.
[20]DU F,BAO X,ZHANG Y,et al.Geocet:Accurate IP geolocation via constraint-based elliptical trajectories[C]//Collaborative Computing:Networking,Applications and Worksharing.2019:603-622.
[21]PATEL K B,MOUKDAD N,ANAND S.Geolocation of iphosts in large computer networks with congestion[C]//ICC 2020-2020 IEEE International Conference on Communications.IEEE,2020:1-6.
[22]KATZ-BASSETT E,JOHN J P,KRISHNAMURTHY A,et al.Towards IP geolocation using delay and topology measurements[C]//Proceedings of the 6th ACM SIGCOMM Conference on Internet Measurement.2006:71-84.
[23]WANG Y,BURGENER D,FLORES M,et al.Towards Street-Level Client-Independent IP Geolocation[C]//8th USENIX Symposium on Networked Systems Design and Implementation.2011:365-379.
[24]KOENKER R,BASSETT G.Regression quantiles[J].Econometrica:Journal of the Econometric Society,1978:46(1):33-50.
[25]VOVK V,GAMMERMAN A,SHAFER G.Algorithmic lear-ning in a random world[M]//Cham:Springer,2005:1-476.
[26]VELIČKOVIĆ P,CUCURULL G,CASANOVA A,et al.Graph attention networks [C]//Proceedings of the International Conference on Learning Representations.2017:1-12.
[27]ROMANO Y,PATTERSON E,CANDES E.Conformalizedquantile regression[J].arXiv:1905.03222,2019.
Viewed
Full text


Abstract

Cited
  Shared   
  Discussed   
No Suggested Reading articles found!
渝ICP备16013121号-4
地址：重庆市两江新区洪湖西路18号    邮编：401121  
电话：023-63500828（编辑部） 023-67039612（会议/专题） 023-67039623（财务/发票）
E-mail：jsjkx12@163.com
本系统由北京玛格泰克科技发展有限公司设计开发