Automatic Extraction Method of Key Information from Logistics Bidding Documents
Logistics bid information identification is an important means for logistics companies to accurately obtain business intelligence.As a source of information,logistics bidding announcements have problems such as irregular text,insufficient label-ing,and difficulty in extracting.Therefore,there is a lack of named entity recognition in the field of logistics bidding documents.To accurately identify the effective information,this study uses logistics bidding announcements obtained from China Logistics Bidding Network and Wutong Network as experimental dataset,then constructs a corpus about logistics bidding documents,and proposes a BERT-BiLSTM-CRF model to identify the tenderee,shipping location and shipping item three entities in the logistics bidding docu-ments.Experimental results show that the model proposed in this paper not only has better entity recognition accuracy but also has strong stability.
国家科技期刊平台
NSTL
万方数据
