Research on acrostic poem generation based on handwritten Chinese character recognition and machine learning

Yanjun Li; Huiran Jia; Yuan Li; Qinglin Wang

doi:10.1109/CAC51589.2020.9327699

Research on acrostic poem generation based on handwritten Chinese character recognition and machine learning

Yanjun Li, Huiran Jia, Yuan Li, Qinglin Wang

School of Automation

Beijing Institute of Technology

Research output: Chapter in Book/Report/Conference proceeding › Conference contribution › peer-review

Abstract

In this paper, we study the automatic generation of acrostic poems based on handwritten Chinese character recognition and machine learning. First, AlexNet network and ResNet50 network are used to train the handwritten Chinese character data set respectively. Select ResNet50 which recognition accuracy is 82.56% as the input module of poem generation model. After this, the GPT-2 model is used to train 26,000 five-character poems in order to build an acrostic poem generation model. Finally, combine these two parts. If we input four images of handwritten Chinese characters, the system can automatically generate a fluent acrostic poem headed by these four characters after character recognition. Experiment results verify the effectiveness of the method.

Original language	English
Title of host publication	Proceedings - 2020 Chinese Automation Congress, CAC 2020
Publisher	Institute of Electrical and Electronics Engineers Inc.
Pages	1287-1292
Number of pages	6
ISBN (Electronic)	9781728176871
DOIs	https://doi.org/10.1109/CAC51589.2020.9327699
Publication status	Published - 6 Nov 2020
Event	2020 Chinese Automation Congress, CAC 2020 - Shanghai, China Duration: 6 Nov 2020 → 8 Nov 2020

Publication series

Name	Proceedings - 2020 Chinese Automation Congress, CAC 2020

Conference

Conference	2020 Chinese Automation Congress, CAC 2020
Country/Territory	China
City	Shanghai
Period	6/11/20 → 8/11/20

Keywords

acrostic poem generation
handwritten Chinese character recognition
machine learning
neural network

Access to Document

10.1109/CAC51589.2020.9327699

Cite this

Li, Y., Jia, H., Li, Y., & Wang, Q. (2020). Research on acrostic poem generation based on handwritten Chinese character recognition and machine learning. In Proceedings - 2020 Chinese Automation Congress, CAC 2020 (pp. 1287-1292). Article 9327699 (Proceedings - 2020 Chinese Automation Congress, CAC 2020). Institute of Electrical and Electronics Engineers Inc.. https://doi.org/10.1109/CAC51589.2020.9327699

@inproceedings{14d2a2b7d3924e79b5d79d9875d7065f,

title = "Research on acrostic poem generation based on handwritten Chinese character recognition and machine learning",

abstract = "In this paper, we study the automatic generation of acrostic poems based on handwritten Chinese character recognition and machine learning. First, AlexNet network and ResNet50 network are used to train the handwritten Chinese character data set respectively. Select ResNet50 which recognition accuracy is 82.56% as the input module of poem generation model. After this, the GPT-2 model is used to train 26,000 five-character poems in order to build an acrostic poem generation model. Finally, combine these two parts. If we input four images of handwritten Chinese characters, the system can automatically generate a fluent acrostic poem headed by these four characters after character recognition. Experiment results verify the effectiveness of the method.",

keywords = "acrostic poem generation, handwritten Chinese character recognition, machine learning, neural network",

author = "Yanjun Li and Huiran Jia and Yuan Li and Qinglin Wang",

note = "Publisher Copyright: {\textcopyright} 2020 IEEE.; 2020 Chinese Automation Congress, CAC 2020 ; Conference date: 06-11-2020 Through 08-11-2020",

year = "2020",

month = nov,

day = "6",

doi = "10.1109/CAC51589.2020.9327699",

language = "English",

series = "Proceedings - 2020 Chinese Automation Congress, CAC 2020",

publisher = "Institute of Electrical and Electronics Engineers Inc.",

pages = "1287--1292",

booktitle = "Proceedings - 2020 Chinese Automation Congress, CAC 2020",

address = "United States",

}

Li, Y, Jia, H, Li, Y & Wang, Q 2020, Research on acrostic poem generation based on handwritten Chinese character recognition and machine learning. in Proceedings - 2020 Chinese Automation Congress, CAC 2020., 9327699, Proceedings - 2020 Chinese Automation Congress, CAC 2020, Institute of Electrical and Electronics Engineers Inc., pp. 1287-1292, 2020 Chinese Automation Congress, CAC 2020, Shanghai, China, 6/11/20. https://doi.org/10.1109/CAC51589.2020.9327699

Research on acrostic poem generation based on handwritten Chinese character recognition and machine learning. / Li, Yanjun; Jia, Huiran; Li, Yuan et al.
Proceedings - 2020 Chinese Automation Congress, CAC 2020. Institute of Electrical and Electronics Engineers Inc., 2020. p. 1287-1292 9327699 (Proceedings - 2020 Chinese Automation Congress, CAC 2020).

Research output: Chapter in Book/Report/Conference proceeding › Conference contribution › peer-review

TY - GEN

T1 - Research on acrostic poem generation based on handwritten Chinese character recognition and machine learning

AU - Li, Yanjun

AU - Jia, Huiran

AU - Li, Yuan

AU - Wang, Qinglin

PY - 2020/11/6

Y1 - 2020/11/6

N2 - In this paper, we study the automatic generation of acrostic poems based on handwritten Chinese character recognition and machine learning. First, AlexNet network and ResNet50 network are used to train the handwritten Chinese character data set respectively. Select ResNet50 which recognition accuracy is 82.56% as the input module of poem generation model. After this, the GPT-2 model is used to train 26,000 five-character poems in order to build an acrostic poem generation model. Finally, combine these two parts. If we input four images of handwritten Chinese characters, the system can automatically generate a fluent acrostic poem headed by these four characters after character recognition. Experiment results verify the effectiveness of the method.

AB - In this paper, we study the automatic generation of acrostic poems based on handwritten Chinese character recognition and machine learning. First, AlexNet network and ResNet50 network are used to train the handwritten Chinese character data set respectively. Select ResNet50 which recognition accuracy is 82.56% as the input module of poem generation model. After this, the GPT-2 model is used to train 26,000 five-character poems in order to build an acrostic poem generation model. Finally, combine these two parts. If we input four images of handwritten Chinese characters, the system can automatically generate a fluent acrostic poem headed by these four characters after character recognition. Experiment results verify the effectiveness of the method.

KW - acrostic poem generation

KW - handwritten Chinese character recognition

KW - machine learning

KW - neural network

UR - http://www.scopus.com/inward/record.url?scp=85100925535&partnerID=8YFLogxK

U2 - 10.1109/CAC51589.2020.9327699

DO - 10.1109/CAC51589.2020.9327699

M3 - Conference contribution

AN - SCOPUS:85100925535

T3 - Proceedings - 2020 Chinese Automation Congress, CAC 2020

SP - 1287

EP - 1292

BT - Proceedings - 2020 Chinese Automation Congress, CAC 2020

PB - Institute of Electrical and Electronics Engineers Inc.

T2 - 2020 Chinese Automation Congress, CAC 2020

Y2 - 6 November 2020 through 8 November 2020

ER -

Li Y, Jia H, Li Y, Wang Q. Research on acrostic poem generation based on handwritten Chinese character recognition and machine learning. In Proceedings - 2020 Chinese Automation Congress, CAC 2020. Institute of Electrical and Electronics Engineers Inc. 2020. p. 1287-1292. 9327699. (Proceedings - 2020 Chinese Automation Congress, CAC 2020). doi: 10.1109/CAC51589.2020.9327699

Research on acrostic poem generation based on handwritten Chinese character recognition and machine learning

Abstract

Publication series

Conference

Keywords

Access to Document

Other files and links

Fingerprint

Cite this