Improving parking availability prediction in smart cities with IoT and ensemble-based model

Smart cities are part of the ongoing advances in technology to provide a better life quality its inhabitants. Urban mobility is one most important components smart cities. Due growing number vehicles these cities, urban traffic congestion becoming more common. In addition, finding places park even car parks not easy for drivers who run circles. Studies have shown that looking parking spaces contribute up 30% congestion. this context, it necessary predict available lots where they want park. We propose paper new system integrates IoT and predictive model based on ensemble methods optimize prediction availability parking. The tests we carried out Birmingham data set allowed reach Mean Absolute Error (MAE) 0.06% average with algorithm Bagging Regression (BR). This results thus improved best existing performance by over 6.6% while dramatically reducing complexity.