IOT-BASED TRAFFIC FLOW PREDICTION IN SMART CITY USING EXTREME GRADIENT BOOSTING ALGORITHM

Abstract

Nowadays, several cities have problems with traffic congestion at certain peak hours, which produces more pollution, noise as well and stress for citizens. The Internet of Things (IoT)-based Intelligent Management System (ITM) is embedded in automatic vehicles and utilized to recognize and transmit data. IoT has been an imminent innovation, moving the globe concerning an automated process as well as (IMS). In this research, the Extreme Gradient Boosting (XGB) is proposed for the prediction of the traffic flow in smart city. XGB is attains the better performance in prediction tasks through the robust handling of complex data structures and relationships. Initially, the data is collected from PeMS07 for estimating the effectiveness of the proposed method. In the pre-processing step, the min-max normalization approach is utilized for the redundancy elimination and minimize data modification errors in the collected dataset. The ResNet101 architecture is utilized to extracting the features from the pre-processed data. The proposed XGB approach attains the MAE of 22.56%, MAPE of 10.46% and RMSE of 35.46% respectively when compared to the existing methods such as DDGCRN.