基于大数据和BiLSTM+CRF的网络空间安全领域命名实体识别研究

Research on Named Entity Recognition in Cyberspace Security Domain Based on Big Data and Bilstm+CRF

由于网络空间安全的知识、技术的更新速度很快,因此很难构建一个非常完备的知识库。基于此,笔者探讨了基于大数据和BiLSTM+CRF的网络空间安全领域命名实体识别,在采集到的网络空间安全文章中抽取出11种网络空间安全领域实体。实验结果表明,该方法的准确率达到98.07%。

Due to the rapid update of cyberspace security knowledge and technology,it is difficult to build a very complete knowledge base.Based on this,the author discussed the recognition of named entities in the cyberspace security domain based on big data and BiLSTM+CRF,and extracted 11 cyberspace security domain entities from the collected cyberspace security articles.Experimental results show that the accuracy of this method reaches 98.07%.