The Application of Neuro-FDT in Urban Short-Term Traffic Flow Prediction

Today, the problems are more and more serious such as urban traffic accident, traffic jam, environment pollution and energy consume etc. Intelligence transportation system is one of valid methods to resolve the problem of road traffic jam and reduce traffic accident, etc. Traffic control and inducement system are the constitution part of the intelligence traffic system. Moreover, how to finish the short-term traffic forecast is becoming premise and key of achieving traffic control and inducement. Therefore, to build up a dependable and accurate forecast model for short-time transportation flows is a popular and core topic in the ITS research realm. In this paper, a new forecast method for city short-term transportation flows is presented. The model based on Neuro-fuzzy decision tree modeling method is different from traditional modeling methods. Neural networks-fuzzy decision tree improves FDT´s classification accuracy and extracts more accuracy human interpretable classification rules. The result of the positive research indicated that this method is very valid for short-term transport flows prediction and it will have a good application prospect in this area.