Title: Error Analysis for Vietnamese Named Entity Recognition on Deep Neural Network Models

Abstract: In recent years, Vietnamese Named Entity Recognition (NER) systems have had a great breakthrough when using Deep Neural Network methods. This paper describes the primary errors of the state-of-the-art NER systems on Vietnamese language. After conducting experiments on BLSTM-CNN-CRF and BLSTM-CRF models with different word embeddings on the Vietnamese NER dataset. This dataset is provided by VLSP in 2016 and used to evaluate most of the current Vietnamese NER systems. We noticed that BLSTM-CNN-CRF gives better results, therefore, we analyze the errors on this model in detail. Our error-analysis results provide us thorough insights in order to increase the performance of NER for the Vietnamese language and improve the quality of the corpus in the future works.