The invention discloses a journey recommendation method based on a knowledge graph and traffic speed prediction, and the method comprises the following steps: A, carrying out the summarization to form an expert experience database, and constructing a multi-level traffic knowledge graph according to the expert experience database; step B, constructing a spatial-temporal road network state topological graph; c, inputting the real-time spatial-temporal road network state topological graph and the multi-level traffic knowledge graph into a traffic speed prediction model to obtain a predicted traffic passing speed, and calculating the passing time of each road section according to the ratio of the road section length to the predicted traffic passing speed; d, generating a corresponding multi-level traffic knowledge graph query statement according to the travel demand of the user, querying the multi-level traffic knowledge graph to obtain a road network sub-graph meeting the requirement, taking the passing time of each road section as the weight of the corresponding edge in the road network sub-graph, and finally generating an optimal travel by using a shortest path algorithm; the method overcomes the problem of time lag, introduces expert knowledge to improve the understanding ability of the algorithm, promotes traffic diversion, and alleviates traffic congestion.
本发明的一种基于知识图谱和交通速度预测的行程推荐方法,其包括如下步骤:步骤A汇总形成专家经验数据库,根据专家经验数据库构建多层级交通知识图谱;步骤B构建时空路网状态拓扑图;步骤C将实时的时空路网状态拓扑图和多层级交通知识图谱输入交通速度预测模型,获得预测交通通行速度,并通过路段长度和预测交通通行速度的比值计算各路段通行时间;步骤D根据用户的出行需求,生成对应的多层级交通知识图谱查询语句,查询多层级交通知识图谱得到符合要求的路网子图,并将各路段通行时间作为路网子图中对应边的权重,最终利用最短路径算法生成最优行程;本发明克服了时滞性问题,引入专家知识提高算法理解能力,促进交通分流,减缓交通拥堵。
Travel recommendation method based on knowledge graph and traffic speed prediction
一种基于知识图谱和交通速度预测的行程推荐方法
2024-06-28
Patent
Electronic Resource
Chinese
Travel time prediction method and device based on knowledge graph
European Patent Office | 2025
|Dynamic graph convolution traffic speed prediction method
European Patent Office | 2020
|Knowledge Graph-Based Enhanced Transformer for Metro Individual Travel Destination Prediction
DOAJ | 2022
|Traffic speed prediction method based on multi-graph cross attention fusion
European Patent Office | 2023
|