Deep Learning in Chinese Text Recognition Program: Exploring the Influence of Program Parameters on Recognition Accuracy

DOI

10.2991/978-94-6463-040-4_110How to use a DOI?

Keywords

Deep learning; Algorithm; Reinforcement learning; Text recognition; Recurrent neural network

Abstract

Deep learning (DL), a class of pattern analysis methods for learning the intrinsic patterns and levels of representation of sample data, has gained more and more attention in recent years. However, deep learning still faces many problems. The difference of various parameters when deep learning is performed for a specific kind of data will have a large impact on the results, while the adjustment of parameters lacks a systematic logical support. Therefore, this paper, through literature research methods, empirical research methods and other relevant methods, will summarize the influencing factors of Chinese text recognition procedures, including the effect of each parameter in the procedure on the results and the differences between different algorithms, and suggest possible directions for future improvements. The paper finds that the choice of algorithm has a large impact on the accuracy of deep learning, and programmers should give priority to choosing the best algorithm rather than obsessing over fine-tuning parameters. The future direction of deep learning should focus on optimizing and improving old algorithms or even creating new, more targeted algorithms that are more efficient for some problems.

Copyright

© 2023 The Author(s)

Open Access

Open Access This chapter is licensed under the terms of the Creative Commons Attribution-NonCommercial 4.0 International License (http://creativecommons.org/licenses/by-nc/4.0/), which permits any noncommercial use, sharing, adaptation, distribution and reproduction in any medium or format, as long as you give appropriate credit to the original author(s) and the source, provide a link to the Creative Commons license and indicate if changes were made.

Download article (PDF)

TY  - CONF
AU  - Zhengyu Peng
PY  - 2022
DA  - 2022/12/27
TI  - Deep Learning in Chinese Text Recognition Program: Exploring the Influence of Program Parameters on Recognition Accuracy
BT  - Proceedings of the 2022 3rd International Conference on Artificial Intelligence and Education (IC-ICAIE 2022)
PB  - Atlantis Press
SP  - 724
EP  - 729
SN  - 2589-4900
UR  - https://doi.org/10.2991/978-94-6463-040-4_110
DO  - 10.2991/978-94-6463-040-4_110
ID  - Peng2022
ER  -