摘要
为提升海洋石油行业工程施工质量检查的智能化水平,提出一种基于中文预训练模型Chinese-BERT和双向长短时记忆网络(BiLSTM)的施工质量问题分类算法。Chinese-BERT模型采用中文预训练能够更好地进行中文文本的特征提取,提高文本向量语义表示的准确性,而BiLSTM则进一步对Chinese-BERT得出的特征向量进行关键信息提取。将模型在真实数据中进行有效性验证,结果显示,所提模型在分类性能上优于其他对比模型,能有效地解决质量问题分类问题。
In order to improve the intelligence level of engineering construction quality inspection in offshore oil industry,a quality problem classification algorithm based on Chinese pre-training model of Chinese-BERT and BiLSTM is proposed.Chinese-BERT model can better extract Chinese text features and improve the accuracy of text vector semantic representation by using Chinese pre-training,while BiLSTM further extracts key information from Chinese-BERT feature vectors.The validity of the model is verified in real data and the results show that the proposed model outperforms other comparative models in terms of classification performance,and it can effectively solve quality problem classification problems.
作者
蔡玉华
安鑫鹏
马永军
安广山
CAI Yuhua;AN Xinpeng;MA Yongjun;AN Guangshan(Bohai Central Station of China Offshore Oil Engineering Quality Supervision,Tianjin 300450,China;College of Artificial Intelligence,Tianjin University of Science and Technology,Tianjin 300457,China)
出处
《现代信息科技》
2025年第7期151-156,共6页
Modern Information Technology
基金
国家重点研发计划(2022YFC2806100)。
