Mathematical Foundations of Computing, volume 5, issue 4, pages 351

CNN models for readability of Chinese texts

Feng Han - ¹

Sizai Hou ²

Le Yin Wei ³

Le-Yin Wei ³

Ding-Xuan Zhou ⁴

Hide authors affiliations

Department of Mathematics, City University of Hong Kong, Hong Kong |

The Hong Kong University of Science and Technology (Guangzhou), Nansha, Guangzhou, 511400, Guangdong, China |

Shaoxing University, Shaoxing 312000, Zhejiang, China |

⁴

School of Data Science, Department of Mathematics, Liu Bie Ju Centre for Mathematical Sciences, City University of Hong Kong, Hong Kong |

Publication type: Journal Article

Publication date: 2022-07-07

American Institute of Mathematical Sciences (AIMS)

Journal: Mathematical Foundations of Computing

scimago Q3

SJR: 0.363

CiteScore: 1.5

Impact factor: 1.3

ISSN: 25778838

DOI: 10.3934/mfc.2022021

Copy DOI

Computational Mathematics

Computational Theory and Mathematics

Artificial Intelligence

Theoretical Computer Science

Abstract

<p style='text-indent:20px;'>Readability of Chinese texts considered in this paper is a multi-class classification problem with <inline-formula><tex-math id="M1">\begin{document}$ 12 $\end{document}</tex-math></inline-formula> grade classes corresponding to <inline-formula><tex-math id="M2">\begin{document}$ 6 $\end{document}</tex-math></inline-formula> grades in primary schools, <inline-formula><tex-math id="M3">\begin{document}$ 3 $\end{document}</tex-math></inline-formula> grades in middle schools, and <inline-formula><tex-math id="M4">\begin{document}$ 3 $\end{document}</tex-math></inline-formula> grades in high schools. A special property of this problem is the strong ambiguity in determining the grades. To overcome the difficulty, a measurement of readability assessment methods used empirically in practice is adjacent accuracy in addition to exact accuracy. In this paper we give mathematical definitions of these concepts in a learning theory framework and compare these two quantities in terms of the ambiguity level of texts. A deep learning algorithm is proposed for readability of Chinese texts, based on convolutional neural networks and a pre-trained BERT model for vector representations of Chinese characters. The proposed CNN model can extract sentence and text features by convolutions of sentence representations with filters and is efficient for readability assessment, which is demonstrated with some numerical experiments.</p>

Found

	1 2 3 4 5 6 7
Mathematical Foundations of Computing	Mathematical Foundations of Computing, 7, 35% Mathematical Foundations of Computing 7 publications, 35%
Journal of Complexity	Journal of Complexity, 2, 10% Journal of Complexity 2 publications, 10%
Analysis and Applications	Analysis and Applications, 1, 5% Analysis and Applications 1 publication, 5%
Advances in Computational Mathematics	Advances in Computational Mathematics, 1, 5% Advances in Computational Mathematics 1 publication, 5%
IEEE Access	IEEE Access, 1, 5% IEEE Access 1 publication, 5%
Journal of Mathematics	Journal of Mathematics, 1, 5% Journal of Mathematics 1 publication, 5%
Foundations of Computational Mathematics	Foundations of Computational Mathematics, 1, 5% Foundations of Computational Mathematics 1 publication, 5%
Journal of Fourier Analysis and Applications	Journal of Fourier Analysis and Applications, 1, 5% Journal of Fourier Analysis and Applications 1 publication, 5%
ACM Transactions on Asian and Low-Resource Language Information Processing	ACM Transactions on Asian and Low-Resource Language Information Processing, 1, 5% ACM Transactions on Asian and Low-Resource Language Information Processing 1 publication, 5%
Neural Networks	Neural Networks, 1, 5% Neural Networks 1 publication, 5%
Neural Computation	Neural Computation, 1, 5% Neural Computation 1 publication, 5%
Communications in Computer and Information Science	Communications in Computer and Information Science, 1, 5% Communications in Computer and Information Science 1 publication, 5%
	1 2 3 4 5 6 7

	1 2 3 4 5 6 7
American Institute of Mathematical Sciences (AIMS)	American Institute of Mathematical Sciences (AIMS), 7, 35% American Institute of Mathematical Sciences (AIMS) 7 publications, 35%
Springer Nature	Springer Nature, 4, 20% Springer Nature 4 publications, 20%
Elsevier	Elsevier, 3, 15% Elsevier 3 publications, 15%
Institute of Electrical and Electronics Engineers (IEEE)	Institute of Electrical and Electronics Engineers (IEEE), 2, 10% Institute of Electrical and Electronics Engineers (IEEE) 2 publications, 10%
World Scientific	World Scientific, 1, 5% World Scientific 1 publication, 5%
Hindawi Limited	Hindawi Limited, 1, 5% Hindawi Limited 1 publication, 5%
Association for Computing Machinery (ACM)	Association for Computing Machinery (ACM), 1, 5% Association for Computing Machinery (ACM) 1 publication, 5%
MIT Press	MIT Press, 1, 5% MIT Press 1 publication, 5%
	1 2 3 4 5 6 7

Are you a researcher?

Create a profile to get free access to personal recommendations for colleagues and new articles.

Metrics

Cite this

GOST | RIS | BibTex | MLA

Found error?

Publisher

American Institute of Mathematical Sciences (AIMS)

Journal

Mathematical Foundations of Computing

scimago Q3

SJR

0.363

CiteScore

1.5

Impact factor

1.3

ISSN

25778838 (Print, Electronic)

CNN models for readability of Chinese texts

Top-30

Journals

Publishers

Are you a researcher?