Encoding multi-granularity structural information for joint Chinese word segmentation and POS tagging

Zhao, Ling; Zhang, Ailian; Liu, Ying; Hao, Fei

doi:10.1016/j.patrec.2020.07.017

Cited by 16 publications

(13 citation statements)

References 3 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…The former is mainly based on word‐based rule matching for a given constructed dictionary, such as positive maximum matching rules, reverse maximum matching rules (Luo et al., 2018) and bidirectional matching rules (Huang et al., 2015; Yunita et al., 2010). The latter is trained on annotated Chinese text to obtain different models: Hidden Markov models (HMMs) and Conditional Random Fields (CRFs), statistical machine learning models (Du et al., 2018; Huang et al., 2017; Liang et al., 2019; Y. Liu et al., 2014; Zhang & Li, 2016), deep learning models (Xu & Sun, 2016; Zhao et al., 2020), etc. Based on the trained model, the text of the unknown label is segmented.…”

Section: Related Workmentioning

confidence: 99%

Chinese Word Segmentation Based on Self‐Learning Model and Geological Knowledge for the Geoscience Domain

Qiu

et al. 2021

Earth and Space Science

View full text Add to dashboard Cite

show abstract

Section: Related Workmentioning

confidence: 99%

Chinese Word Segmentation Based on Self‐Learning Model and Geological Knowledge for the Geoscience Domain

Qiu

et al. 2021

Earth and Space Science

View full text Add to dashboard Cite

show abstract

“…Shao proposed a bidirectional RNN-CRF architecture that incorporated rich contextual information and sub-character level features [8]. Zhao presented a model based on lattice-LSTM and Convolutional Network, exploiting character, word, and subword information [9]. Tian proposed a two-way attention neural network model using context features and their corresponding syntactic information of characters in the sequence [10].…”

Section: Related Workmentioning

confidence: 99%

“…It is a character-based model and utilizes rich contextual information and sub-character level features [8]. Zhao presented a lattice-LSTM and Convolutional Network, which can exploit multi-granularity of information, including characters, words, and subwords [9]. Tian introduced a neural network model with a two-way attention mechanism.…”

Section: Introductionmentioning

confidence: 99%

Untitled

2021

International Journal of Advanced Research in Computer and Comm

View full text Add to dashboard Cite

Recently, deep learning methods have greatly improved the state-of-the-art in many natural language processing tasks. Previous work shows that the Transformer can capture long-distance relations between words in a sequence. In this paper, we propose a Transformer-based neural model for Chinese word segmentation and part-ofspeech tagging. In the model, we present a word boundary-based character embedding method to overcome the character ambiguity problem. After the Transformer layer, BiLSTM-CRF layer is used to generate the best tagging results. Experiments on Chinese Treebank show that our model on Chinese word segmentation and part-of-speech tagging outperforms the baseline model and achieves state-of-the-art performance.

show abstract

“…Park et al [15] proposed regularization of subwords, using a unary language model to generate multiple candidate subword sequences, enriching the input of the encoder to enhance the robustness of the translation system. Zhao et al [16] introduced the representation of multigranularity BPE to obtain the semantic representation of vocabulary on average. Zhang et al [17] believed that the encoder word vector layer, decoder word vector layer, and decoder output layer have different functions, so the choice of BPE granularity for different layers should also be different.…”

Section: Introductionmentioning

confidence: 99%

Neural Network Machine Translation Method Based on Unsupervised Domain Adaptation

Wang

2020

Complexity

View full text Add to dashboard Cite

Relying on large-scale parallel corpora, neural machine translation has achieved great success in certain language pairs. However, the acquisition of high-quality parallel corpus is one of the main difficulties in machine translation research. In order to solve this problem, this paper proposes unsupervised domain adaptive neural network machine translation. This method can be trained using only two unrelated monolingual corpora and obtain a good translation result. This article first measures the matching degree of translation rules by adding relevant subject information to the translation rules and dynamically calculating the similarity between each translation rule and the document to be translated during the decoding process. Secondly, through the joint training of multiple training tasks, the source language can learn useful semantic and structural information from the monolingual corpus of a third language that is not parallel to the current two languages during the process of translation into the target language. Experimental results show that better results can be obtained than traditional statistical machine translation.

show abstract

Encoding multi-granularity structural information for joint Chinese word segmentation and POS tagging

Cited by 16 publications

References 3 publications

Chinese Word Segmentation Based on Self‐Learning Model and Geological Knowledge for the Geoscience Domain

Chinese Word Segmentation Based on Self‐Learning Model and Geological Knowledge for the Geoscience Domain

Untitled

Neural Network Machine Translation Method Based on Unsupervised Domain Adaptation

Contact Info

Product

Resources

About