Pregledni rad
https://doi.org/10.7307/ptt.v27i6.1762
Analysed potential of big data and supervised machine learning techniques in effectively forecasting travel times from fused data
Ivana Šemanjski
; Faculty of transport and traffic sciences, University of Zagreb, Vukeliceva 4, 1000 Zagreb, Croatia
Sažetak
Travel time forecasting is an interesting topic for many ITS services. Increased availability of data collection sensors increases the availability of the predictor variables but also highlights the high processing issues related to this big data availability. In this paper we aimed to analyse the potential of big data and supervised machine learning techniques in effectively forecasting travel times. For this purpose we used fused data from three data sources (Global Positioning System vehicles tracks, road network infrastructure data and meteorological data) and four machine learning techniques (k-nearest neighbours, support vector machines, boosting trees and random forest).
To evaluate the forecasting results we compared them in-between different road classes in the context of absolute values, measured in minutes, and the mean squared percentage error. For the road classes with the high average speed and long road segments, machine learning techniques forecasted travel times with small relative error, while for the road classes with the small average speeds and segment lengths this was a more demanding task. All three data sources were proven itself to have a high impact on the travel time forecast accuracy and the best results (taking into account all road classes) were achieved for the k-nearest neighbours and random forest techniques.
Ključne riječi
big data; support vector machines; k-nearest neighbours; boosting trees; random forest; forecasting travel times; data fusion
Hrčak ID:
151925
URI
Datum izdavanja:
17.12.2015.
Posjeta: 1.090 *