Designing Machine Learning Driven Dual Adjacency Graph Based Spatiotemporal Traffic Prediction for Smarter Urban Mobility and Congestion Managemnet in Bengaluru City

Abstract With the mushrooming inflation in the population, transportation systems are challenged by several issues. Traffic congestion is customary and traffic accidents occurs often repeatedly deteriorating traffic environments. To take the edge of these issues and enhance the transportation efficiency, accurate traffic forecasting is critical. Accurate temporal time-dependent traffic predictions are essential for ensuring safety and efficiency of intelligent traffic management system. Nevertheless, owing to the intrinsic spatial and temporal dependencies of traffic flow, it is still a challenging problem. To solve this, some methods are proposed taking into consideration the detailed traffic patterns across major roads and intersections, while the complicated spatiotemporal dynamics and interdependencies between traffic flows are not taken into account.In this work a method called, Gaussian Dual Adjacency Graph-based Spatial Correlated and Temporal Time-dependent (GDAG-SCTT) traffic prediction in Bangalore city is proposed. Initially with the raw traffic patterns obtained from Bangalore’s traffic pulse dataset as input are subjected to Local-Global Invariant Inter Quartile and Min-Max Normalization based Traffic Data Pre-processing. By applying this pre-processing outliers are removed and finally normalized pre-processed results are obtained. Next, extraction of spatial and temporal features is done by using Gaussian Kernel Dynamic Adjacency based Spatial Correlated and Temporal Time-dependency based feature extraction model. Here, first, Spatial Correlated Graph Convolutional Neural Network is applied to extract spatial features, following which Temporal Long Short Term Time-dependency Memory is applied to extract temporal features. To evaluate the GDAG-SCTT methods performance, five different performance metrics, precision, recall, accuracy, root mean square error and training time are validated and analyzed. The GDAG-SCTT achieved higher performance compared to other state-of-the-art methods on our collected Bangalore’s traffic pulse dataset demonstrating the efficiency in reducing root mean square error by 28% while improving overall accuracy by 25% in an extensive manner.