Road traffic prediction based on base station location data by Random Forest

Now road traffic monitoring is very meaningful to the serious road traffic congestion problem all over the world. Operators have a large number of base stations location information (OIDD) which can be used to detect road condition. However, for the reason that base station data have the disadvantage of inaccurate location, fixed location and so on, direct calculating the road evaluation speed is not effective. In this paper, by extracting diverse features from base station data, we propose new method based on machine learning model Random Forest and classification association to predict road condition. We apply our method to real complex roads in some province. Take the existing commercial road app real-time monitoring as the benchmark, the experiments demonstrate that our method performs better than the traditional road prediction algorithms.