Jinming Zhong

Title: Innovations by Jinming Zhong in Traffic Anomaly Detection

Introduction

Jinming Zhong is a notable inventor based in Guangzhou, China. He has made significant contributions to the field of traffic anomaly detection through his innovative methods and systems. His work focuses on enhancing the accuracy and stability of traffic data analysis.

Latest Patents

Jinming Zhong holds a patent for a "Traffic anomaly detection method and system based on improved BERT integrating contrastive learning." This method involves several steps, including obtaining and preprocessing traffic data, building an improved BERT model with an embedding layer and 12 Transformer encoder networks, and performing weight sharing among the encoder networks. The process also includes building a classification network and a total loss function based on cross-entropy and contrastive loss. The model undergoes unsupervised pre-training and fine-tuning, ultimately leading to a trained model that can accurately detect traffic anomalies.

Career Highlights

Jinming Zhong is affiliated with Jinan University, where he continues to advance his research in traffic anomaly detection. His work is characterized by a commitment to improving the generalization ability of models while ensuring stability and accuracy in traffic data analysis.

Collaborations

Jinming Zhong collaborates with colleagues such as Feiran Huang and Zhibo Zhou, contributing to a dynamic research environment focused on innovative solutions in traffic management.

Conclusion

Jinming Zhong's contributions to traffic anomaly detection exemplify the impact of innovative thinking in addressing real-world challenges. His patent and ongoing research at Jinan University highlight the importance of advancements in technology for improving traffic systems.