Multi-scale Semantic Collaborative Power Grid Fault Text Classification Model Combined with ChineseBERT

To address the problems of incomplete feature extraction of the traditional model and low semantic representation of static word vector, a power grid fault text classification model based on ChineseBERT-BiSRU-MCNN-AT is proposed. The ChineseBERT integrates font and pinyin information, combines word context information for dynamic learning and solves polysemy. The BiSRU-MCNN extracts the text’s local semantic and global sequence features to ensure the extracted features’ comprehensiveness. Soft attention gives the model the ability to recognize keywords. Experiments on power fault texts show that ChineseBERT-BiSRU-MCNN-AT achieves the highest F1 value.