| 169 | 0 | 376 |
| 下载次数 | 被引频次 | 阅读次数 |
道路网选取需要集成多种类型不同层次的道路实体上下文特征，涉及复杂的决策过程。近年来，机器学习方法的运用不断提升道路网选取模型的性能。与此同时，选取模型的复杂度也逐步增加，导致可解释问题凸显，继而影响模型的可信度和实际应用价值。针对这一问题，本文提出了一种可解释的道路网选取机器学习方法。具体地，采用XGBoost构建道路选取模型，同时引入SHAP(SHapley Additive exPlanations)方法解析不同道路特征对选取结果的影响。实验结果表明，SHAP方法有助于揭示模型的决策逻辑，从而增强模型的可解释性。
Abstract：Road network selection requires to integrate various features of road entities at different levels, involving a complex decision-making process. In recent years, the application of machine learning methods has continuously improved the performance of road network selection models. However, the increasing complexity of these models has also brought explainability problem, which affects the model's credibility and practical application value. To address this challenge, this study explores the interpretable machine learning methods for road network selection. Specifically, a road selection model is constructed by using XGBoost and incorporates SHAP(SHapley Additive exPlanations) to analyze the influence of different road features on the selection outcomes. The experimental results demonstrate that the SHAP method helps uncover the decision-making logic of the road network selection model, thereby enhancing its explainability.
[1] T?PFER F,PILLEWIZER W.The principles of selection[J].The Cartographic Journal,1966,3(1):10-16.
[2] HARRIEL E.The constraint method for solving spatial conflicts in cartographic generalization[J].Cartography and Geographic Information Science,1999,26(1):55-69.
[3] SHEA K S,MCMASTER R B.Cartographic generalization in a digital environment:when and how to generalize[C]//Auto-carto IX:Proceedings of the International Symposium on Computer-Assisted Cartography.Baltimore,USA,1989:56-67.
[4] LIU X,ZHAN F B,AI T H.Road selection based on Voronoi diagrams and “strokes” in map generalization[J].International Journal of Applied Earth Observation and Geoinformation,2010,12:S194-S202.
[5] 杨敏，艾廷华，周启.顾及道路目标stroke特征保持的路网自动综合方法[J].测绘学报，2013,42(4):581-587.YANG M,AI T H,ZHOU Q.A method of road network generalization considering stroke properties of road object[J].Acta Geodaetica et Cartographica Sinica,2013,42(4):581-587.
[6] 钟东，郭庆胜，林青，等.基于蚁群算法的道路选取模型研究[J].测绘工程，2017,26(4):47-52.ZHONG D,GUO Q S,LIN Q,et al.Research of ant colony algorithm model of road selection[J].Engineering of Surveying and Mapping,2017,26(4):47-52.
[7] KARSZNIA I,SIELICKA K,WEIBEL R.Optimising road selection for small-scale maps using decision tree-based models[C]//AutoCarto 23rd International Research Symposium on Cartography and GIScience Cartography and Geographic Information Society.Redlands,USA,2020:37-42.
[8] 刘凯，李进，沈婕，等.基于BP神经网络和拓扑参数的道路网选取研究[J].测绘科学技术学报，2016,33(3):325-330.LIU K,LI J,SHEN J,et al.Selection of road network using BP neural network and topological parameters[J].Journal of Geomatics Science and Technology,2016,33(3):325-330.
[9] ZHOU Q,LI Z.A comparative study of various supervised learning approaches to selective omission in a road network[J].The Cartographic Journal,2017,54(3):254-264.
[10] TOUYA G,ZHANG X,LOKHAT I.Is deep learning the new agent for map generalization?[J].International Journal of Cartography,2019,5(2/3):142-157.
[11] KANG Y,GAO S,ROTH R E.Artificial intelligence studies in cartography:a review and synthesis of methods,applications,and ethics[J].Cartography and Geographic Information Science,2024,51(4):599-630.
[12] JEPSEN T S,JENSEN C S,NIELSEN T D.Graph convolutional networks for road networks[C]//Proceedings of the 27th ACM SIGSPATIAL International Conference on Advances in Geographic Information Systems.Chicago,USA,2019:460-463.
[13] GUO X,LIU J,WU F,et al.A method for intelligent road network selection based on graph neural network[J].ISPRS International Journal of Geo-Information,2023,12(8):336.
[14] TANG J,DENG M,PENG J,et al.Automatic road network selection method considering functional semantic features of roads with graph convolutional networks[J].International Journal of Geographical Information Science,2024,38(11):2403-2432.
[15] ZHENG H,ZHANG J,LI H,et al.Road network intelligent selection method based on heterogeneous graph attention neural network[J].ISPRS International Journal of Geo-Information,2024,13(9):300.
[16] BARREDO ARRIETA A,DíAZ-RODRíGUEZ N,DEL SER J,et al.Explainable Artificial Intelligence (XAI):concepts,taxonomies,opportunities and challenges toward responsible AI[J].Information fusion,2020,58:82-115.
[17] YUAN Y,GUO W,TANG S,et al.Effects of patterns of urban green-blue landscape on carbon sequestration using XGBoost-SHAP model[J].Journal of Cleaner Production,2024,476:143640.
[18] LUI A,LAMB G W.Artificial intelligence and augmented intelligence collaboration:regaining trust and confidence in the financial sector[J].Information & Communications Technology Law,2018,27(3):267-283.
[19] KHAN S,GHAZAL T M,ALYAS T,et al.Towards transparent traffic solutions:reinforcement learning and explainable AI for traffic congestion[J].International Journal of Advanced Computer Science and Applications,2025,16(1):503-511.
[20] LI H,HAN Z,SUN Y,et al.CGMega:explainable graph neural network framework with attention mechanisms for cancer gene module dissection[J].Nature Communications,2024,15:5997.
[21] MERSHA M,LAM K,WOOD J,et al.Explainable artificial intelligence:a survey of needs,techniques,applications,and future direction[J].Neurocomputing,2024,599:128111.
[22] AGARWAL R,MELNICK L,FROSST N,et al.Neural additive models:interpretable machine learning with neural nets[J].Advances in Neural Information Processing Systems,2021,34:4699-4711.
[23] HINTON G,VINYALS O,DEAN J.Distilling the knowledge in a neural network[DB/OL].(2015-03-09)[2025-08-27].https://arxiv.org/abs/1503.02531.
[24] GOU J,YU B,MAYBANK S J,et al.Knowledge distillation:a survey[J].International Journal of Computer Vision,2021,129(6):1789-1819.
[25] RIBEIRO M T,SINGH S,GUESTRIN C.“Why should I trust you?” explaining the predictions of any classifier[C].The 22nd ACM SIGKDD International Conference on Knowledge Discovery and Data Mining.San Francisco,USA,2016:1135-1144.
[26] LUNDBERG S M,LEE S I.A unified approach to interpreting model predictions[C]//The 31st International Conference on Neural Information Processing Systems.Long Beach,USA,2017:4768-4777.
[27] SHRIKUMAR A,GREENSIDE P,KUNDAJE A.Learning important features through propagating activation differences[C]//The 34th International Conference on Machine Learning.Sydney,Australia,2017:3145-3153.
[28] PRAKASH E I,SHRIKUMAR A,KUNDAJE A.Towards more realistic simulated datasets for benchmarking deep learning models in regulatory genomics[C]//Proceedings of the 16th Machine Learning in Computational Biology Meeting.Online,2022:58-77.
[29] CHEN T,GUESTRINC.XGBoost:a scalable tree boosting system[C]//Proceedings of the 22nd ACM SIGKDD International Conference on Knowledge Discovery and Data Mining.San Francisco,USA,2016:785-794.
[30] THOMSON R C,RICHARDSON D E.The “good continuation” principle of perceptual organization applied to the generalization of road networks[C]//Proceedings of the ICA 19th International Cartographic Conference.Ottawa,Canada,1999:1215-1223.
[31] THOMSONR C.The ‘stroke’concept in geographic network generalization and analysis[C]//Progress in Spatial Data Handling:12th International Symposium on Spatial Data Handling.Vienna,Austria,2006:681-697.
[32] ZHOU Q,LI Z.A comparative study of various strategies to concatenate road segments into strokes for map generalization[J].International Journal of Geographical Information Science,2012,26(4):691-715.
[33] PORTA S,CRUCITTI P,LATORA V.The network analysis of urban streets:a dual approach[J].Physica A:Statistical Mechanics and its Applications,2006,369(2):853-866.
[34] YING R,BOURGEOIS D,YOU J,et al.GNNExplainer:generating explanations for graph neural networks[J].Advances in Neural Information Processing Systems,2019,32:9240-9251.
基本信息:
中图分类号:TP181;P208
引用信息:
[1]吕翔,徐晓,杨敏.道路网选取的可解释机器学习方法[J].测绘科学技术学报,2025,41(05):486-493.
基金信息:
智慧地球重点实验室基金项目(KF2023ZD04-01)
2025-05-27
2025
2025-08-24
2025-11-03
2025
1
2025-11-26
2025-11-26
2025-11-26