关键词: 命名实体识别, 嵌套语义, 多尺度注意力, 卷积神经网络, 子空间

Abstract: The accuracy of named entity recognition task will promote the research of multiple downstream tasks in natural language field. Due to a large number of nested semantics in text, named entities are recognized difficultly. Recognizing nested semantics becomes a difficulty in natural language processing. Previous studies have single scale of extracting feature and under-utilization of the boundary information. They ignore many details under different scales and then lead to the situation of entity recognition error or omission. Aiming at the above problems, a named entity recognition method based on multi-scale attention is proposed. First, the BERT model is used to obtain representation vector containing context information, and then the BiLSTM network is used to strengthen the context representation of text. Second, the representation vectors are enumerated and concatenated to form span information matrix. The direction information is fused to obtain richer interactive information. Third, multi-head attention is used to construct multiple subspaces. Two-dimensional convolution is used to optionally aggregate text information at different scales in each subspace, so as to implement multi-scale feature fusion in each attention layer. Finally, the fused matrix is used for span classification to identify named entities. The experimental results show that the F1 score of the proposed method reaches 81.7% and 86.8% on GENIA and ACE2005 English datasets, respectively. The proposed method is better than the existing mainstream models.

Key words: named entity recognition, nested semantic, multi-scale attention, convolutional neural network, subspace