Abstract-Real-time and accurate short-term traffic flow prediction is a key issue and difficult in traffic control and guidance. Using data mining and large data-driven principle, nonparametric regression is a better method to resolve shortterm traffic flow prediction. But there are two main obstacles that case base is difficult to be generated and search is slow. For this reason, this paper presents a short-term traffic flow prediction method based on balanced binary tree and K-NEAREST NEIGHBOR NONPARAMETRIC REGRESSION (KNN2NPR). Case base is generated through clustering method and balance binary tree structure. K-nearest neighbor nonparametric regression improves accuracy of prediction and fulfills the real-time requirement. The prediction example in this paper demonstrates that this method is effective.