Named Entity Recognition Model Based on Feature Fusion

Named entity recognition can deeply explore semantic features and enhance the ability of vector representation text data. This paper proposes a named method based on multi-head attention to aim at problem fuzzy lexical boundary in Chinese recognition. Firstly, Word2vec is used extract word vectors, HMM ALBERT character Feedforward-attention mechanism fuse three then fused vectors remove by BiLSTM. Then mine potential information features. Finally, label classification results are output after conditional random field screening. Through verification WeiboNER, MSRA, CLUENER2020 datasets, show that proposed algorithm effectively improve performance