关键词: 命名实体识别, 序列标注, 边界偏移, 交并比, 边界调整

Abstract: To address the issue of positional deviation between the predicted entity boundaries and the true entity boundaries in sequence labeling models for Named Entity Recognition (NER), a sequence labeling optimization method combined entity boundary offset was proposed. Firstly, the concept of boundary offset was introduced to quantify the positional relationship between each word and the entity boundaries. The relative offset between each word to the nearest entity boundary was calculated, and these offsets are used to generate candidate spans for the entity boundaries. Subsequently, the Intersection-over-Union (IoU) was used as a filtering criterion to exclude low-quality candidate spans, retaining those most likely to represent the entity boundary. Finally, the boundary adjustment module was used to update the positions of entity boundaries in the label sequence based on the candidate spans, thereby optimizing the entity boundaries in the entire label sequence and improving the performance of entity recognition. The method achieves F1-scores of 80.48%, 96.42%, and 94.80% on the CLUENER2020, Resume-zh, and MSRA datasets, respectively, validating its effectiveness for the named entity recognition task.

Key words: named entity recognition, sequence labeling, boundary offset, intersection-over-union, boundary adjustment