In order to develop knowledge graphs to help experts in the vulnerability industry to quickly grasp vulnerability information, we need an entity recognition model to automatically extract entities in the vulnerability domain from massive text data. Considering that entities in the vulnerability domain are a mixture of Chinese and English, this paper proposes an entity recognition model, ERNIE + BILSTM + CRF, which combines the ERNIE pre-training model with the BILSTM and CRF models to enhance the extraction of semantic information from Chinese text. Experimental results show the effectiveness efficiency of the model in entity recognition for both Chinese and English texts.