This study presents a systematic process combining trajfic forecasting and data mining models for traffic information systems. Fuzzy c-means clustering model was developed for mining traffic flow-speed-occupancy relationships, then to extrapolate traffic information. The hybrid grey-based recurrent neural network (G-RNN) was developed for traffic parameter forecasting. G-RNN integrates grey modeling into recurrent neural networks that is capable of dealing with both randomness and spatial-temporal properties in trajfic data implicitly. Field data from Taiwan national freeway was used as an example for testing the proposed models. Study results were shown that the G-RNN model is capable of predicting traffic parameters with a high degree of accuracy. The application presents three clusters built from data and recognized three types of traffic conditions. Study results also showed feasibility of the method for advanced traffic information systems.
2005 IEEE International Conference on Information Reuse and Integration, p.p 356-361