首页|基于深度学习的交通运输行业数据自动分级方法研究

基于深度学习的交通运输行业数据自动分级方法研究

扫码查看
为促进交通运输行业信息系统互联、保障数据安全,进而推动行业健康发展,本文对交通运输行业中的数据共享和自动分级方法进行研究.本文基于卷积神经网络(convolutional neural networks,CNN)-双向门控循环单元(bidirectional gating recurrent unit,BiGRU)-胶囊网络(capsule network,CapsNet)模型的数据类别判定方法,设计并实现了交通运输信息资源目录系统中数据的自动分级模块,完成了行业领域下的数据自动分级.实验结果表明本文算法的准确率和F1 值分别达到了70.48%和70.16%,明显高于现有的几种主流模型,可以有效提高数据分级的效率.
Research on automatic classification method of transportation industry data based on deep learning
To promote the interconnection of information systems in the transportation industry,ensure data security,and further promote healthy development of the industry,this paper studies data sharing and automatic classification methods in the transportation industry.Based on the data classification method of convolutional neural network-bidirectional gating recurrent unit-capsule network(CNN-BiGRU-CapsNet)model,this paper designs and implements the automatic data classification module in the transportation information resource directory system,and completes the automatic classification of data in the industry field.The experimental results show that the accuracy and F1 score of the proposed algorithm have reached 70.48%and 70.16%respectively,which are significantly higher than several existing mainstream models and can effectively improve the efficiency of data classification.

transportation industrydata sharingdata securitydata classificationdeep learningconvolutional neural networkbidirectional gating recurrent unitcapsule network

王继晔、张少博、叶润泽、张绍阳

展开 >

陕西省交通运行监测中心,陕西西安 710075

长安大学信息工程学院,陕西西安 710064

西安电子科技大学党委组织部,陕西西安 710071

交通运输行业 数据共享 数据安全 数据分级 深度学习 卷积神经网络 双向门控循环单元 胶囊网络

陕西省交通厅科技项目

20-15X

2024

应用科技
哈尔滨工程大学

应用科技

CSTPCD
影响因子:0.693
ISSN:1009-671X
年,卷(期):2024.51(2)
  • 15