Text Classification Model of Power Equipment Defects Based on Chinese BERT and Multi Feature Cooperative Network
To address the problems of incomplete feature extraction and inaccurate semantic expression of word vector in tradi-tional models,a text classification model of power equipment defects based on ChineseBERT and multi feature collaborative network is proposed.ChineseBERT model optimized for Chinese characters is used to extract text vector representation to im-prove the accuracy of word vector semantic representation.Multi feature collaborative network comprehensively captures the local and contextual semantic features of defective text.The soft attention mechanism improves the ability of the model to focus on key features.Experiments on real power equipment defect text data show that the classification performance of the model is better than the recent deep learning model,and the F1 score is as high as 96.82％,which proves the effectiveness of the mod-el.
text classificationChineseBERTmulti feature collaborationsoft attention
万方数据
维普
