Research on Classification Method of Construction Laws and Regulations Data DOI
Chunkai Wang,

Bianping su,

Yusong Wang

et al.

Lecture notes in computer science, Journal Year: 2024, Volume and Issue: unknown, P. 258 - 268

Published: Jan. 1, 2024

Language: Английский

Research on multi-label short text categorization method for online education under deep learning DOI Open Access

Yinuo Guo

Applied Mathematics and Nonlinear Sciences, Journal Year: 2025, Volume and Issue: 10(1)

Published: Jan. 1, 2025

Abstract The rapid development of the modern Internet has not only changed our way life, but also previous mode education and learning, online been greatly developed improved accordingly. In this paper, BERT model is used to extract word vectors multilabel short texts for education, then BiLSTM-CNN features texts, a classifier constructed by Sigmoid activation function realize output classification results texts. validation analysis model’s effectiveness was conducted using public dataset THCNEWS self-collected EduData as examples. loss Marco-P after 5*10 5 steps training converged stably around 0.085 vs. 96.05%. Marco-F1 values multi-label text on datasets reach 0.915 0.962, which are significantly higher than individual comparison models. Combining deep learning technology with can achieve accurate data provide new exploration ideas improving quality education.

Language: Английский

Citations

0
A survey of large language model-augmented knowledge graphs for advanced complex product design DOI

Xinxin Liang,

Zuoxu Wang, Jihong Liu

et al.

Journal of Manufacturing Systems, Journal Year: 2025, Volume and Issue: 80, P. 883 - 901

Published: April 29, 2025

Language: Английский

Citations

0
Research of multi-label text classification based on label attention and correlation networks DOI Creative Commons
Ling Yuan, Xinyi Xu, Ping Sun

et al.

PLoS ONE, Journal Year: 2024, Volume and Issue: 19(9), P. e0311305 - e0311305

Published: Sept. 30, 2024

Multi-Label Text Classification (MLTC) is a crucial task in natural language processing. Compared to single-label text classification, MLTC more challenging due its vast collection of labels which include extracting local semantic information, learning label correlations, and solving data imbalance problems. This paper proposes model Label Attention Correlation Networks (LACN) address the challenges classifying multi-label enhance classification performance. The proposed employs attention mechanism for discriminative representation uses correlation network based on distribution results. Also, weight factor number samples modulation function prediction probability are combined alleviate effectively. Extensive experiments conducted widely-used conventional datasets AAPD RCV1-v2, extreme EUR-LEX AmazonCat-13K. results indicate that can be used deal with achieve optimal or suboptimal versus state-of-the-art methods. For dataset, compared method, it outperforms second-best method by 2.05% ∼ 5.07% precision@k 2.10% 3.24% NDCG@k k = 1, 3, 5. superior outcomes demonstrate effectiveness LACN competitiveness dealing tasks.

Language: Английский

Citations

2
Enhanced automated text categorization via Aquila optimizer with deep learning for Arabic news articles DOI Creative Commons
Muhammad Swaileh A. Alzaidi, Alya Alshammari, Abdulkhaleq Q. A. Hassan

et al.

Ain Shams Engineering Journal, Journal Year: 2024, Volume and Issue: unknown, P. 103189 - 103189

Published: Nov. 1, 2024

Language: Английский

Citations

1
Discovering Granger causality with convolutional neural networks DOI

Oktay Sahinoglu,

Ayça Kumluca Topallı, İhsan Topallı

et al.

Journal of Intelligent Manufacturing, Journal Year: 2024, Volume and Issue: unknown

Published: Dec. 1, 2024

Language: Английский

Citations

1
Research on Classification Method of Construction Laws and Regulations Data DOI
Chunkai Wang,

Bianping su,

Yusong Wang

et al.

Lecture notes in computer science, Journal Year: 2024, Volume and Issue: unknown, P. 258 - 268

Published: Jan. 1, 2024

Language: Английский

Citations

0