A Review of Three Types of Neural Networks: Convolutional, Recurrent, and Recursive
Main Article Content
Keywords
convolutional neural networks, recurrent neural networks, deep learning, network structure, feature extraction, temporal data, model fusion
Abstract
Artificial neural networks are a core technology in the field of nonlinear system modeling and intelligentization. Convolutional, Recurrent, and Recurrent Neural Networks, as their three core branches, each have unique structural and performance characteristics, and related research urgently needs systematic review and integration. This article adopts the methods of literature combing and comparative analysis to comprehensively sort out the development vein of the three types of neural networks, analyze their structural characteristics and applicable scenarios, explore the technical pain points and optimization paths of various networks, and look forward to the future development direction. The innovation of this research lies in the research results of the development and application of the three types of neural networks in the system, clearly defining the technical characteristics, differences between advantages and disadvantages and integration application modes of each network, and clarifying their respective optimization directions. The study found that the three types of neural networks have their own advantages and disadvantages, complementary adaptation, and have irreplaceable application value in computer vision, natural language processing and other fields. In the future, its development needs to focus on optimizing its own technical defects, improving the ability of model generalization, further expand the application boundaries through deep integration with other network models and learning mechanisms, and promote the application of intelligent technology in complex scenarios.
References
- [1] Zhou, F., Jin, L., & Dong, J. (2017). A review of convolutional neural networks. Chinese Journal of Computers, 40(6), 1229–1251.
- [2] Li, Y., Hao, Z., & Lei, H. (2016). A review of convolutional neural networks. Journal of Computer Applications, 36(9), 2508–2515, 2565.
- [3] Yang, L., Wu, Y., Wang, J., & Zhou, C. (2018). A review of recurrent neural networks. Journal of Computer Applications, 38(S2), 1–6, 26.
- [4] Liang, C., Liu, Y., Zhou, J., & Li, J. (2021). Multi-point wind speed forecasting method in wind farms based on convolutional recurrent neural network. Power System Technology, 45(2). https://doi.org/10.13335/j.1000-3673.pst.2020.0767
- [5] Gu, Z., Chen, C., Zheng, J., & Wu, H. (2022). Traffic flow prediction based on spatio-temporal graph convolutional recurrent neural network. Control and Decision, 37(3). https://doi.org/10.13195/j.kzyjc.2020.1445
- [6] Zhang, L., Liu, S., & Tian, Y. (2021). Traffic state index prediction based on the fusion of convolutional and recurrent neural networks. Traffic and Transportation, 37(1).
- [7] Wang, Y., Liao, B., Peng, C., & Yang, X. (2021). A review of recurrent neural networks. Journal of Jishou University (Natural Sciences Edition), 42(1). https://doi.org/10.13438/j.cnki.jdzk.2021.01.007
- [8] Liu, W., Liang, X., & Qu, H. (2016). Research on learning performance of convolutional neural networks with different pooling models. Journal of Image and Graphics, 21(9).
- [9] Yan, C., & Wang, C. (2021). Development and application of convolutional neural network models. Journal of Frontiers of Computer Science and Technology, 15(1).
- [10] Gao, M., & Xu, B. (2019). Recommendation algorithm based on recurrent neural network. Computer Engineering, 45(8). https://doi.org/10.19678/j.issn.1000-3428.0051294
- [11] Wang, Y. (2025). Research on neural network models for forest fire risk prediction[Master's thesis, Tianjin University of Technology]. https://doi.org/10.27360/d.cnki.gtlgy.2024.000605
- [12] Cong, S., & Dai, Y. (2004). Research on the structure of recurrent neural networks. Journal of Computer Applications, (8), 18–20, 27.
- [13] Gao, X., Zhu, H., Zhao, K., & Wang, J. (2026). Lithology identification method based on convolutional neural network-long short-term memory artificial neural network-self-attention mechanism. Science Technology and Engineering, 26(2).
- [14] Zhu, H., Zhen, Q., Chen, Q., & Wang, H. (2026). A novel mental fatigue detection method based on the fusion of single-channel EEG convolutional neural network and transformer. Journal of Biomedical Engineering.
- [15] Wang, K. (2025). Research on multispectral image compression algorithm based on convolutional recurrent neural network[Master's thesis, Nanjing University of Aeronautics and Astronautics]. https://doi.org/10.27239/d.cnki.gnhhu.2023.003084
Article Sidebar
This work is licensed under a Creative Commons Attribution 4.0 International License.