Relationship extraction of red character resources based on attention mechanism deep neural networks

In this study, historical resources about the red historical figure Zhu Rui were collected through field surveys, online research, and expert interviews. To create a domain ontology suitable for Zhu Rui, the figure's characteristics and resource types were taken into consideration. Through semantic annotation and classification of the ontology, entity categories related to red historical figures' resources were identified. To improve accuracy, relationship extraction was performed using a convolutional neural network model with an attention mechanism. This study provides a foundation for a semantic understanding of resources related to red historical figures, which is essential for digitalizing and providing knowledge services. However, limitations exist due to the diverse types of resources and subjective views. Future work will focus on optimizing entity extraction to further improve relationship extraction accuracy.