Research on university email analysis based on SVM-RFE and Transformer-TBAM

LI Zhen; LI Zhichao; CHEN Lin

doi:10.11959/j.issn.1000-436x.2024229

您当前的位置：

首页 >

文章列表页 >

Research on university email analysis based on SVM-RFE and Transformer-TBAM

Campus Network Construction | 更新时间：2024-12-31

- Research on university email analysis based on SVM-RFE and Transformer-TBAM
- Journal on Communications Vol. 45, Issue Z2, Pages: 97-101(2024)
- 作者机构：
  
  1.山东大学信息化工作办公室，山东济南 250100
  2.山东大学（威海）信息化工作办公室，山东威海 265209
- 作者简介：
- 基金信息：
- DOI：10.11959/j.issn.1000-436x.2024229
  CLC： TP391
- Received：22 October 2024，
  
  Published：30 November 2024
- 稿件说明：
移动端阅览
李振,李智超,陈琳.基于SVM-RFE与Transformer-TBAM的高校邮件分析研究[J].通信学报,2024,45(Z2):97-101.

LI Zhen,LI Zhichao,CHEN Lin.Research on university email analysis based on SVM-RFE and Transformer-TBAM[J].Journal on Communications,2024,45(Z2):97-101.
李振,李智超,陈琳.基于SVM-RFE与Transformer-TBAM的高校邮件分析研究[J].通信学报,2024,45(Z2):97-101. DOI： 10.11959/j.issn.1000-436x.2024229.

LI Zhen,LI Zhichao,CHEN Lin.Research on university email analysis based on SVM-RFE and Transformer-TBAM[J].Journal on Communications,2024,45(Z2):97-101. DOI： 10.11959/j.issn.1000-436x.2024229.

摘要

通过挖掘高校电子邮件文本数据并进行分析，可以帮助教职工更好地了解学生的意见和建议，提高管理效率。目前，深度学习方法是文本情感分析的主要方法，然而现有的方法没有充分利用中文文本中的特征。为解决此问题，提出基于SVM-RFE与Transformer-TBAM架构模型处理高校邮件，该架构重构了双通道注意力模型及特征筛选机制以深度提取有效特征信息。实验表明，该算法在高校邮件数据集分类效果达到了94.67%的准确率，比传统算法高出1.2%。

Abstract

By mining and analyzing email text data from universities

it can help faculty members better understand students’opinions and suggestions

and improve management efficiency. At present

deep learning methods are the main approach for text sentiment analysis

but existing methods have not fully utilized the features in Chinese text. To address this issue

a framework based on SVM-RFE and Transformer models was proposed for processing university emails. This architecture reconstructs a dual branch attention model and feature filtering mechanism to deeply extract effective feature information. The experiment shows that the algorithm achieves an accuracy of 94.67% in the classification of university email datasets

which is 1.2% higher than traditional algorithms.

关键词

Keywords

references

孟祥福 , 石皓源 . 基于Transformer模型的时序数据预测方法综述 [J/OL ] . 计算机科学与探索 , ( 2024-07-30 )[ 2024-10-22 ] .

闫芳序 , 王剑辉 . 基于SVM和Word2Vec的微博评论情感识别模型 [J ] . 现代计算机 , 2024 , 30 ( 10 ): 60 - 64 .

YAN F X , WANG J H . Sentiment recognition model of weibo comments based on SVM and Word2Vec [J ] . Modern Computer , 2024 , 30 ( 10 ): 60 - 64 .

PENNINGTON J , SOCHER R , MANNING C . Glove: global vectors for word representation [C ] // Proceedings of the 2014 Conference on Empirical Methods in Natural Language Processing . Stroudsburg : ACL Press , 2014 : 1532 - 1543 .

LECUN Y , BOTTOU L , BENGIO Y , et al . Gradient-based learning applied to document recognition [J ] . Proceedings of the IEEE , 1998 , 86 ( 11 ): 2278 - 2324 .

ZAREMBA W , SUTSKEVER I , VINYALS O . Recurrent Neural Network Regularization [J ] . arXiv Preprint , arXiv: 1409.2329 , 2014 .

HOCHREITER S , SCHMIDHUBER J . Long short-term memory [J ] . Neural Computation , 1997 , 9 : 1735 - 1780 .

UMER M , IMTIAZ Z , AHMAD M , et al . Impact of convolutional neural network and FastText embedding on text classification [J ] . Multimedia Tools and Applications , 2023 , 82 ( 4 ): 5569 - 5585 .

BETUL POLAT S , CANKURT S . Fake news classification using BLSTM with glove embedding [C ] // Proceedings of the 2023 17th International Conference on Electronics Computer and Computation (ICECCO) . Piscataway : IEEE Press , 2023 : 1 - 5 .

LI X L , ZHANG Y Y , JIN J , et al . A model of integrating convolution and BiGRU dual-channel mechanism for Chinese medical text classifications [J ] . PLoS One , 2023 , 18 ( 3 ): 1 - 20 .

YOON K . Convolutional neural networks for sentence classification [C ] // Proceedings of the 2014 Conference on Empirical Methods in Natural Language Processing . Stroudsburg : ACL Press , 1746 - 1751 .

LAI S W , XU L H , LIU K , et al . Recurrent convolutional neural networks for text classification [C ] // Proceedings of the AAAI Conference on Artificial Intelligence , 2015 , 29 ( 1 ): 2267 - 2273 .

Views

958

下载量

CSCD

Alert me when the article has been cited

提交

Tools

Publicity Resources

Illicit data resale detection method via an enhanced relational graph convolutional network

Network traffic anomaly detection method based on multi-scale convolution and channel attention mechanism

Trajectory reconstruction attacks on differential privacy based on a CNN-BiLSTM-Attention hybrid model

Modulation recognition method based on multiscale convolutional fusion coding networks

THz UM-MIMO system channel estimation algorithm based on deep residual block fixed-point network

Related Author

Wang Yuxiang

Zhang Lingcui

Hou Yuqiao

Yang Qian

Niu Ben

Fu Yu

Wang Yujue

Yu Yihan

Related Institution

State Key Laboratory of Cyberspace Security Defense

School of Cyber Security, University of Chinese Academy of Sciences

Institute of Information Engineering, Chinese Academy of Sciences

Department of Operational Operations and Planning, Naval University of Engineering

Department of Information Security, Naval University of Engineering

AI问答

⁰