Research Article
Online Traffic Accident Spatial-Temporal Post-Impact Prediction Model on Highways Based on Spiking Neural Networks
Algorithm 2
Multi-ReSuMe-based classifier.
(1) | Definition | (2) | Historic accident dataset | (3) | accident data in where | (4) | the minimum value of network error to reach when learning is considered converged | (5) | synaptic weight between all neurons | (6) | maximum number of iterations | (7) | Initialization | (8) | Initial synaptic weight with random value | (9) | For iteration 1, M do | (10) | For do | (11) | Encode to a series of spike trains | (12) | Set membrane potential of all neurons to the resting potential (set to 0) | (13) | Pass input spike trains to the network and find actual output spike | (14) | Calculate network error | (15) | Compute weight modifications for all layers according to equation (2) and equation (4) | (16) | Update synaptic weight | (17) | End for | (18) | Calculate summed network error | (19) | While |
|