Research on Chinese Named Entity Recognition Based on ResNet Algorithm

View at publisher Summarise with VIRAT
Authors: Xi Liu
Cited by: 0
References: 10

Summary

No summary yet. VIRAT can read this paper and write a plain-language summary of its aim, methods, key findings and limitations.

Summarise with VIRAT Free to try with your Science Catalogs account.

Abstract

This paper proposes a convolutional neural network optimization strategy for detecting the dispersion state of crushed stones in synchronous crushed stone seals. The study constructed a dataset of gravel scattering states and analyzed the high linear correlation between image channels using the Pearson correlation coefficient. Subsequently, deep convolution was adopted instead of standard convolution to optimize the ResNet series models. The experimental results show that the computational complexity of the optimized ResNet50, ResNet101 and ResNet152 models is significantly reduced, and the average cross-union ratio is improved, verifying the effectiveness of the optimization strategy. Meanwhile, the research also attempted to combine the self-attention mechanism with IDCNN and apply it to the recognition of named entities in judicial documents, achieving good recognition results. This paper not only provides an efficient method for detecting the dispersion state of crushed stones, but also offers new ideas for the field of named entity recognition.

Fields of Study & Subjects

Advanced Neural Network Applications Topic Modeling Infrastructure Maintenance and Monitoring

Publication details

Journal / source
IEEE International Conference on Power, Electronics and Green Energy (ICPEGE)
Pages
1211–1217
Publication date
28 January 2026
Language
EN