Multi-modal IoT data fusion for real-time sports event analysis and decision support

With the rapid development of Internet of Things (IoT) technology, real-time sports event analysis and decision support face challenges arising from the integration of multi-source spatiotemporal data. To address this issue, this paper proposes a real-time sports event analysis and decision support model based on multi-modal IoT data fusion—ST-TransBay. This model combines Spatiotemporal Graph Convolutional Networks (ST-GCN), Transformer, and a Bayesian optimization module to enhance the efficiency and accuracy of decision support systems in dynamic sports environments. To validate the effectiveness of the model, we conducted experiments using the UCI HAR and WISDM datasets. Evaluation metrics include accuracy, recall, F1-score, and inference time. The experimental results show that ST-TransBay achieved an accuracy of 95.4% and 94.6% on the two datasets, with inference times as low as 5.2 ms and 6.1 ms, respectively. These results demonstrate the model’s potential in real-time sports event analysis and provide valuable insights for decision support systems in related fields.