基于Chinese-BERT和BiLSTM的海洋石油工程行业质量问题分类

打开文本图片集

中图分类号：TP3191.1；TP18 文献标识码：A 文章编号：2096-4706（2025）07-0151-06

Abstract： Inorder toimprove theintelligencelevelofengineering construction quality inspection inofshoreoilindustry a quality problem clasification algorithm basedon Chinese pre-training modelof Chinese-BERTandBiLSTMis proposed. Chinese-BERTmodelcanbeter extract Chinese textfeaturesandimprovetheaccuracyoftextvectorsemanticrepresentationby using Chinese pre-training，whileBiLSTMfurtherextractskey informationfrom Chinese-BERTfeature vectors.Thevalidityof the modelis verified inrealdataandtheresults showthat the proposed modeloutperformsothercomparativemodels interms of classification performance，and it can effectively solve quality problem classfication problems.

eywords：offshore oil industry;quality problem; text classification; Chinese-BERT; BiLSTN

0 引言

海洋石油工程行业中，工程施工质量检查具有至关重要的地位和作用，通过施工质量检查获取的质检数据是衡量与判断工程项目质量的重要标准和技术依据，高效的项目质量检测不仅能够提高项目的施工质量以及可维护性。（剩余8579字）