Short-Term Prediction of Vehicle Speed on Main City Roads using the k-Nearest Neighbor Algorithm

Abstract

Road speed is an important measure in transportation. It can be employed in various ways including traffic congestion detection, travel time estimation, and road design. Consequently, accurate speed prediction is essential in the development of intelligent transportation systems. In this paper we present an analysis and speed prediction of a certain road section in Busan, South Korea. In previous works, only historical data of the target road are used for prediction. Here, we extract features from the real traffic data by taking the neighboring roads into consideration. After obtaining the candidate features, linear regression, model tree, and k-nearest neighbor (k-NN) are employed for both feature selection and speed prediction. Experiment result shows that k-NN outperforms model tree and linear regression for the given dataset. Compared to the other predictors, k-NN significantly reduces the error measures that we use, including mean absolute percentage error (MAPE) and root mean square error (RMSE).