Rajšp Alen, Fister Iztok
Faculty of Electrical Engineering and Computer Science, University of Maribor, Koroška cesta 46, Maribor SI-2000, Slovenia.
Data Brief. 2023 May 27;48:109251. doi: 10.1016/j.dib.2023.109251. eCollection 2023 Jun.
Navigating through a real-world map can be represented in a bi-directed graph with a group of nodes representing the intersections and edges representing the roads between them. In cycling, we can plan training as a group of nodes and edges the athlete must cover. Optimizing routes using artificial intelligence is a well-studied phenomenon. Much work has been done on finding the quickest and shortest paths between two points. In cycling, the solution is not necessarily the shortest and quickest path. However, the optimum path is the one where a cyclist covers the suitable distance, ascent, and descent based on his/her training parameters. This paper presents a Neo4j graph-based dataset of cycling routes in Slovenia. It consists of 152,659 nodes representing individual road intersections and 410,922 edges representing the roads between them. The dataset allows the researchers to develop and optimize cycling training generation algorithms, where distance, ascent, descent, and road type are considered.
在现实世界的地图中导航可以用一个双向图来表示，其中一组节点代表十字路口，边代表它们之间的道路。在自行车运动中，我们可以将训练规划为运动员必须经过的一组节点和边。利用人工智能优化路线是一个经过充分研究的现象。在寻找两点之间最快和最短路径方面已经做了很多工作。在自行车运动中，解决方案不一定是最短和最快的路径。然而，最佳路径是骑自行车的人根据他/她的训练参数覆盖合适的距离、爬坡和下坡的路径。本文展示了一个基于Neo4j图的斯洛文尼亚自行车路线数据集。它由152,659个代表各个道路交叉口的节点和410,922条代表它们之间道路的边组成。该数据集使研究人员能够开发和优化考虑距离、爬坡、下坡和道路类型的自行车训练生成算法。
