TY - JOUR
T1 - Interpreting RandomlyWired GraphModels for Chinese NER
AU - Chen, Jie
AU - Xu, Jiabao
AU - Xi, Xuefeng
AU - Cui, Zhiming
AU - Sheng, Victor S.
N1 - Publisher Copyright:
© 2023 Tech Science Press. All rights reserved.
PY - 2023
Y1 - 2023
N2 - Interpreting deep neural networks is of great importance to understand and verify deep models for natural language processing (NLP) tasks. However, most existing approaches only focus on improving the performance of models but ignore their interpretability. In this work, we propose a Randomly Wired Graph Neural Network (RWGNN) by using graph to model the structure of Neural Network, which could solve two major problems (word-boundary ambiguity and polysemy) of ChineseNER. Besides, we develop a pipeline to explain the RWGNNby using Saliency Map and Adversarial Attacks. Experimental results demonstrate that our approach can identify meaningful and reasonable interpretations for hidden states of RWGNN.
AB - Interpreting deep neural networks is of great importance to understand and verify deep models for natural language processing (NLP) tasks. However, most existing approaches only focus on improving the performance of models but ignore their interpretability. In this work, we propose a Randomly Wired Graph Neural Network (RWGNN) by using graph to model the structure of Neural Network, which could solve two major problems (word-boundary ambiguity and polysemy) of ChineseNER. Besides, we develop a pipeline to explain the RWGNNby using Saliency Map and Adversarial Attacks. Experimental results demonstrate that our approach can identify meaningful and reasonable interpretations for hidden states of RWGNN.
KW - Named entity recognition
KW - graph neural network
KW - interpretation
KW - random graph network
KW - saliency map
UR - http://www.scopus.com/inward/record.url?scp=85138816880&partnerID=8YFLogxK
U2 - 10.32604/cmes.2022.020771
DO - 10.32604/cmes.2022.020771
M3 - Article
AN - SCOPUS:85138816880
SN - 1526-1492
VL - 134
SP - 747
EP - 761
JO - CMES - Computer Modeling in Engineering and Sciences
JF - CMES - Computer Modeling in Engineering and Sciences
IS - 1
ER -