A Comparison of Artificial Intelligence-Based Methods in Traffic Prediction

Traffic plays an important role in our society as its state can affect individuals and industries in various ways. Traffic congestion can bring negative impacts to the society and can lead to bigger problems if let be without a solution to mitigate it. Thus, traffic prediction serves as a solution to said problem. In this systematic literature review, AI-based traffic prediction methods are compared in order to find which ones serve as the better solutions for predicting traffic. Using the PRISMA Flowchart methodology, which helps authors systematically analyze relevant publications and improve the quality of reports and meta-analyses. By conducting further analysis on the screened references, it is found that the methods that integrates Convolutional Neural Network or Recurrent Neural Network with Long Short-Term Memory along with error-recurrent Neural Network proved to be good candidates for an optimal traffic prediction.