Article: Economic monitoring and early warning based on feature screening and hybrid neural network Journal: International Journal of Information and Communication Technology (IJICT) 2025 Vol.26 No.21 pp.23 - 38 Abstract: Intending to the issue that the existing study do not fully exploit features, the random forest algorithm (IRF) is improved first. The splitting feature screening function is simplified based on the principle of infinitesimal equivalence, and the Gini coefficient value of the non-category attribute is introduced to improve the computational efficiency of the algorithm. Then, public health economic impact variables are selected, and spatial features are extracted using a residual convolutional neural network. Temporal features are extracted using a gate rate unit (GRU), and a self-attention mechanism is incorporated to enhance the spatial and temporal features. Finally, the IRF filter is used to select the most important spatio-temporal features of the early warning results and map them to the monitoring and early warning results through nonlinear transformation. The experimental outcome indicates that the accuracy of the proposed model has been improved by 5.07%-14.85%. Inderscience Publishers - linking academia, business and industry through research

Title: Economic monitoring and early warning based on feature screening and hybrid neural network

Authors: Dongfang Dai

Addresses: College of Electronic Commerce, Tangshan University, Tangshang 063000, China

Abstract: Intending to the issue that the existing study do not fully exploit features, the random forest algorithm (IRF) is improved first. The splitting feature screening function is simplified based on the principle of infinitesimal equivalence, and the Gini coefficient value of the non-category attribute is introduced to improve the computational efficiency of the algorithm. Then, public health economic impact variables are selected, and spatial features are extracted using a residual convolutional neural network. Temporal features are extracted using a gate rate unit (GRU), and a self-attention mechanism is incorporated to enhance the spatial and temporal features. Finally, the IRF filter is used to select the most important spatio-temporal features of the early warning results and map them to the monitoring and early warning results through nonlinear transformation. The experimental outcome indicates that the accuracy of the proposed model has been improved by 5.07%-14.85%.

Keywords: economic early warning; random forest; feature screening; residual convolutional neural network; gate rate unit; GRU.

DOI: 10.1504/IJICT.2025.146833

International Journal of Information and Communication Technology, 2025 Vol.26 No.21, pp.23 - 38

Received: 15 Apr 2025
Accepted: 29 Apr 2025
Published online: 20 Jun 2025 *

Title: Economic monitoring and early warning based on feature screening and hybrid neural network

Keep up-to-date