Named entity identification method fusing entity boundary information

The invention relates to a named entity recognition method fusing entity boundary information, which comprises the following steps of: 1) inputting a Chinese text into a BERT model to obtain Chinese word embedding information; 2) inputting word embedding information into a left boundary information extraction network model, firstly capturing a long-distance dependency relationship through BiLSTM, and processing a label imbalance problem and learning boundary information by using a Focal Loss loss function; 3) inputting word embedding information into a right character feature extraction network model, and learning more character features through combination of an attention layer and a BiLSTM layer; and 4) the CRF model is used for learning the dependency relationship between the tags, the accuracy and robustness of the model are improved, and the method is more suitable for Chinese named entity recognition tasks. According to the method, the accuracy and robustness of Chinese entity extraction can be improved