Journal of Geo-information Science >
Hyperparameter Selection for Urban Metro Travel Knowledge Graph Embedding
Received date: 2023-02-10
Revised date: 2023-04-19
Online published: 2023-06-02
Supported by
National Natural Science Foundation of China(42171449)
National Natural Science Foundation of China(42101464)
Open Research Fund Program of LIESMARS(21I03)
Knowledge graphs are an important data infrastructure in AI technologies and applications, and have become a hot research topic in geosciences. The size and topological features in geographic knowledge graphs are usually different from universal knowledge graphs, which are not typical small-world networks. However, existing studies often use the default network search depth when learning geographic knowledge graph representations, and its rationality needs further demonstration. For this purpose, this paper constructs a metro travel knowledge graph based on the topological structure features of metro line network, combined with passenger flow data, POI (Point of Interest) data and built environment data, etc.; then GraphSAGE model is used to learn node multidimensional feature embedding and combine POI data for semantic recognition of station classification results to verify the suitable network search depth for metro travel knowledge graph. The results showed that, compared to the default 2 layers search depth, the node embedding features of this metro travel knowledge graph work optimally when the search depth is 3 layers. This study shows that the hyperparameter selection of the geographic knowledge graph representation is supposed to take into account the geographic features, and it is important to avoid the use of results from fields such as computer science that have not been distinguished. When the search depth is 3 layers, the metro station classification results are also more reasonable and explanatory, which can provide a basis for station planning and passenger flow prediction using knowledge graph and AI methods.
LUO Qiuyu , YUE Yang , GU Yanyan . Hyperparameter Selection for Urban Metro Travel Knowledge Graph Embedding[J]. Journal of Geo-information Science, 2023 , 25(6) : 1164 -1175 . DOI: 10.12082/dqxxkx.2023.230054
表1 依据城市用地类型的POI数据的重分类Tab. 1 Reclassification of POI data by urban land use type |
大类 | 中类 |
---|---|
居住 | 第一类居住用地;第二类居住用地; 第三类居住用地;第四类居住用地 |
公共服务 | 公共设施;科教文化;体育休闲;医疗保健;政府机构 |
商业 | 餐饮服务;购物服务;金融保险;汽车摩托;生活服务;住宿服务 |
办公 | 公司企业;商务写字楼 |
交通 | 道路附属;地址地名;交通设施 |
绿地与广场 | 风景名胜;公园广场 |
表2 4种不同类型的居住用地Tab. 2 Four different types of residential land uses |
居住用地类型 | 内容 |
---|---|
一类居住用地 | 独立式住宅(别墅),配套齐全,布局完整 |
二类居住用地 | 以多层、中高层为主,配套齐全,布局完整 |
三类居住用地 | 单身宿舍 |
四类居住用地 | 原农村居民住宅用地 |
表3 地铁刷卡数据示例Tab. 3 Examples of smart card data |
ID | 时间 | 线路 | 进出站 | 名称 | 经度/°E | 纬度/°N |
---|---|---|---|---|---|---|
68580**** | 2018-11-05T20:49:29 | 地铁一号线 | 3(进站) | 站A | 114.**** | 22.**** |
68580**** | 2018-11-05T21:30:16 | 地铁四号线 | 4(出站) | 站B | 114.**** | 22.**** |
[1] |
张雪英, 张春菊, 吴明光, 等. 顾及时空特征的地理知识图谱构建方法[J]. 中国科学:信息科学, 2020, 50(7):1019-1032.
[
|
[2] |
诸云强, 孙凯, 胡修棉, 等. 大规模地球科学知识图谱构建与共享应用框架研究与实践[J]. 地球信息科学学报, 2022.
[
|
[3] |
刘峤, 李杨, 段宏, 等. 知识图谱构建技术综述[J]. 计算机研究与发展, 2016, 53(3):582-600.
[
|
[4] |
艾瑞咨询产业数字化研究部人工智能研究组. 一图胜万言,一目了然中国知识图谱行业研究报告[C]// 艾瑞咨询系列研究报告, 2022(8):323-400.
[ Artificial Intelligence Research Group, Industry Digitalization Research Department, iResearch Consulting. A picture is worth a thousand words, a glance at Chinese knowledge graph industry research report[C]// iResearch Consulting Research Report Series, 2022(8):323-400. ] DOI:10.26914/c.cnkihy.2022.035148
|
[5] |
王益鹏, 张雪英, 党玉龙, 等. 顾及时空过程的台风灾害事件知识图谱表示方法[J]. 地球信息科学学报, 2022.
[
|
[6] |
王志华, 杨晓梅, 周成虎. 面向遥感大数据的地学知识图谱构想[J]. 地球信息科学学报, 2021, 23(1):16-28.
[
|
[7] |
杨玉基, 许斌, 胡家威, 等. 一种准确而高效的领域知识图谱构建方法[J]. 软件学报, 2018, 29(10):2931-2947.
[
|
[8] |
蒋秉川, 万刚, 许剑, 等. 多源异构数据的大规模地理知识图谱构建[J]. 测绘学报, 2018, 47(8):1051-1061.
[
|
[9] |
陆锋, 余丽, 仇培元. 论地理知识图谱[J]. 地球信息科学学报, 2017, 19(6):723-734.
[
|
[10] |
高嘉良, 仇培元, 余丽, 等. 基于旅游知识图谱的可解释景点推荐[J]. 中国科学:信息科学, 2020, 50(7):1055-1068.
[
|
[11] |
|
[12] |
刘俊楠, 刘海砚, 陈晓慧, 等. 面向多源地理空间数据的知识图谱构建[J]. 地球信息科学学报, 2020, 22(7):1476-1486.
[
|
[13] |
|
[14] |
|
[15] |
|
[16] |
|
[17] |
|
[18] |
|
[19] |
|
[20] |
|
[21] |
方滨兴, 贾焰, 韩毅. 社交网络分析核心科学问题、研究现状及未来展望[J]. 中国科学院院刊, 2015, 30(2):187-199.
[
|
[22] |
吕鹏辉, 张士靖. 学科知识网络研究(Ⅰ)引文网络的结构、特征与演化[J]. 情报学报, 2014, 33(4):340-348.
[
|
[23] |
|
[24] |
|
[25] |
|
[26] |
|
[27] |
|
[28] |
|
[29] |
|
[30] |
侯秀芳, 冯晨, 左超, 等. 2022年中国内地城市轨道交通线路概况[J]. 都市快轨交通, 2023, 36(1):9-13.
[
|
[31] |
|
[32] |
|
[33] |
|
[34] |
|
[35] |
方勇. 手机信令数据在轨道站点客流与土地利用关联性研究中的应用——以工作日早高峰进站客流为例[D]. 深圳: 深圳大学, 2017.
[
|
[36] |
谷岩岩, 焦利民, 董婷, 等. 基于多源数据的城市功能区识别及相互作用分析[J]. 武汉大学学报·信息科学版, 2018, 43(7):1113-1121.
[
|
[37] |
|
[38] |
|
[39] |
|
[40] |
|
[41] |
|
[42] |
|
[43] |
|
[44] |
|
/
〈 |
|
〉 |