Improving person name recognition quality in Chinese text with reinforced processing of ambiguities

In this paper, we focus on the task of person name recognition (PNR) in Chinese text, which is an important part of Named entity recognition (NER). Chinese word segmentation makes the sequence tagging of PNR more accurate than character-based PNR, but it also brings more word-level ambiguities. For reducing the negative effect of word segmentation for PNR, we reinforce the analysis of ambiguities and propose a model to deal with them. Experimental results on People´s Daily corpus show that the proposed model is effective in PNR.