VnCoreNLP: A Vietnamese Natural Language Processing Toolkit

Abstract

We present an easy-to-use and fast toolkit, namely VnCoreNLP-a Java NLP annotation pipeline for Vietnamese. Our VnCoreNLP supports key natural language processing (NLP) tasks including word segmentation, part-of-speech (POS) tagging, named entity recognition (NER) and dependency parsing, and obtains state-of-the-art (SOTA) results for these tasks. We release VnCoreNLP to provide rich linguistic annotations to facilitate research work on Vietnamese NLP. Our VnCoreNLP is open-source and available at: https:// github.com/vncorenlp/VnCoreNLP.

Attentive Neural Network for Named Entity Recognition in Vietnamese

“…We then make use of the information of word, those generated POS, and chunking tags to train VNER model. Moreover, we also evaluate VNER model on VLSP-2016 dataset with another setup as in the experiment of VnCoreNLP ( [11]) in which the contiguous syllable constituting a PER tag is merged to form a word. In comparison to other neural network for Vietnamese NER task, we compare the performance of VNER model with two neural models: NNVLP model ( [8]) that makes use of the combination of bidirectional LSTM, CNN, and CRF models; and vie-ner-lstm model ( [14]) that incorporates automatic syntactic features with word embeddings as input for bidirectional LSTM network.…”

“…There have been considerable work proposed by Vietnamese researchers in solving the NER problem such as dynamic feature induction model ( [11]), CRF model ( [2]), or LSTM ( [8], [9]). CRF-based model achieves state-of-the-art results on the VLSP 2016 and VLSP 2018 competitions; however, it still suffers from the linear statistical model drawbacks as mentioned above.…”

“…Due to VLSP-2016 dataset that does not have development set, hence we create a development set by sampling randomly 2000 samples of train set as in [11]; and the rest of train set is used for training VNER model. We then train VNER model on both VLSP-2016 and VLSP-2018 datasets with the train set and further tune up the model with the development set.…”

BERT+vnKG: Using Deep Learning and Knowledge Graph to Improve Vietnamese Question Answering System

“…Finally, we create a relationship for the triple (e i , v i , e i+1 ). Besides that, we also experienced another library named VnCoreNLP [23]. VnCoreNLP is also an open source that can be downloaded from [23].…”

“…Besides that, we also experienced another library named VnCoreNLP [23]. VnCoreNLP is also an open source that can be downloaded from [23]. The difference with underthesea is that VnCoreNLP has dependency parsing feature that can analyze relationships between parts of speech in a Vietnamese sentence.…”