OPTIMIZATION OF DYNAMIC NEURAL NETWORKS PERFORMANCE FOR SHORT-TERM TRAFFIC PREDICTION
Sherif Ishak, Prashanth Kotha, Ciprian Alecsandru · Transportation Research Board 82nd Annual MeetingTransportation Research Board · 2003
This paper presents an approach to optimize the short-term traffic prediction performance using multiple topologies of dynamic neural networks and various network-related and traffic-related settings. The study emphasizes the potential benefit of optimizing the prediction performance by deploying multi-model approaches under parameters and traffic condition settings. The emphasis of the paper is on the application of temporal-processing topologies in short-term speed predictions in the range of 5 to 20 minute-horizons. Three network topologies are utilized: Jordan/Elman, partially recurrent networks and time-lagged feedforward networks. The input patterns were constructed from data collected at the target location, as well as the upstream and downstream locations. However, various combinations were also considered. To encourage the networks to associate with historical information on recurrent conditions, a time factor was attached to the input patterns to introduce time recognition capabilities, in addition to information encoded in the recent past data. The optimal prediction settings (type of topology and input settings) were determined such that the performance is maximized under different traffic conditions at the target and adjacent locations. The optimized performance of the dynamic neural networks was compared to that of a statistical non-linear time series approach, which was outperformed in most cases