Time related association rule mining with Accuracy Validation in traffic volume prediction with large scale simulator

Genetic Network Programming (GNP) based time related association rules mining method provides an useful mean to investigate future traffic volumes of road networks and hence helps to develop traffic navigation systems. Further improvements have been proposed in this paper about the time related association rule mining using generalized GNP with Accuracy Validation. For better adapting to the real-time traffic situations of the large scale simulator, the mechanism of Accuracy Validation is studied. The aim of this algorithm is to better handle association rule extraction using prediction accuracy as criteria and guide the whole evolution process. The generalized algorithm which can find the important time related association rules is described and experimental results are presented considering a traffic prediction problem using the database provided by a large scale simulator SOUND/4U.