A Spatio-Temporal Neural Network Approach for Predicting Traffic Accident Hotspot Areas: A Case Study in Bandung City

Traffic accidents are the most significant cause of human death after disease, with the majority of victims being of productive age. In order to decrease the number of traffic accidents, it is crucial to identify where and when they occur most frequently. The Spatio-Temporal Neural Network (STNN) approach is a reasonably new yet superior method for modeling space-time data. However, this method needs to be improved to deal with the limitations of spatial datasets related to traffic accidents in Bandung. The primary focus of this study is to enhance the performance of STNN in producing forecasts regarding traffic accident hotspots. Several experiments have been carried out, including conducting experiments with several zoom levels, comparing STNN with traditional machine learning methods, and applying GetisOrd Gi* to the dataset. The results showed that using the GetisOrd Gi* approach provides an additional advantage in improving the predictive result of STNN.