Please use this identifier to cite or link to this item:
|Title:||A machine learning-based approach to time-dependent shortest path queries||Authors:||Wei, Yumou||Keywords:||DRNTU::Engineering::Computer science and engineering||Issue Date:||2017||Abstract:||Road traffic is known to be time-dependent. The travel time of a road varies at different times of the day. Many algorithms have been proposed for finding a shortest path in a time-dependent road network. In this project, I explored an alternative approach that leveraged on GPS trajectories collected from thousands of taxis. Each GPS trajectory was mapped to a set of real road segments. An abstract landmark graph was built to represent the city’s road network and a machine learning-based approach was proposed to estimate the travel time of each edge. The estimates made by this approach were compared against real-time estimates made by existing online mapping services to evaluate its accuracy. A modified Dijkstra’s algorithm was presented to calculate a shortest path in a time-dependent landmark graph, based on the travel time estimates.||URI:||http://hdl.handle.net/10356/70473||Rights:||Nanyang Technological University||Fulltext Permission:||restricted||Fulltext Availability:||With Fulltext|
|Appears in Collections:||SCSE Student Reports (FYP/IA/PA/PI)|
Items in DR-NTU are protected by copyright, with all rights reserved, unless otherwise indicated.