<abstract xmlns="http://www.w3.org/1999/xhtml">

<p>In this paper, we aim to progress the efficiency and accuracy of information processing and security detection in computer networks by introducing convolutional neural networks in machine learning algorithms that are capable of multi-scaling from the channel attention module and spatial attention module in extracting image information. Global maximum pooling and global averaging are done for the feature maps generated by both modules to get the clearest feature maps by dimensionality reduction. The loss function is used to calculate the feature maps to reduce the data loss generated during data extraction and finally complete the image data processing. To verify the effectiveness of the proposed platform, network images containing different amounts of data are input into the platform, and the accuracy and loss of data extraction are obtained. The results show that the data extraction accuracy of the reduced platform is up to 100%, which is 6% higher than other platforms. The number of data losses in other platforms is more than twice of this paper, while the number of losses in this paper can be controlled within 5. It can be seen that convolutional neural network in machine learning improves the accuracy of data extraction from computer network information images and reduces the loss in data extraction.</p>
</abstract>

In this paper, we aim to progress the efficiency and accuracy of information processing and security detection in computer networks by introducing convolutional neural networks in machine learning algorithms that are capable of multi-scaling from the channel attention module and spatial attention module in extracting image information. Global maximum pooling and global averaging are done for the feature maps generated by both modules to get the clearest feature maps by dimensionality reduction. The loss function is used to calculate the feature maps to reduce the data loss generated during data extraction and finally complete the image data processing. To verify the effectiveness of the proposed platform, network images containing different amounts of data are input into the platform, and the accuracy and loss of data extraction are obtained. The results show that the data extraction accuracy of the reduced platform is up to 100%, which is 6% higher than other platforms. The number of data losses in other platforms is more than twice of this paper, while the number of losses in this paper can be controlled within 5. It can be seen that convolutional neural network in machine learning improves the accuracy of data extraction from computer network information images and reduces the loss in data extraction.

Research on computer network information image data processing based on machine learning

Applied Mathematics and Nonlinear Sciences

This work is licensed under the Creative Commons Attribution 4.0 International License.

In this paper, we aim to progress the efficiency and accuracy of information processing and security detection in computer networks by introducing...

Research on computer network information image data processing based on machine learning

Online veröffentlicht: 07. Okt. 2023

Seitenbereich: -

Eingereicht: 06. Jan. 2023

Akzeptiert: 17. Apr. 2023

DOI: https://doi.org/10.2478/amns.2023.2.00504

Schlüsselwörter
Convolutional neural network, Machine learning, Maximum global pooling, Global averaging, Loss function, Image data processing

© 2023 Haifeng Mei et al., published by Sciendo

This work is licensed under the Creative Commons Attribution 4.0 International License.

Research on computer network information image data processing based on machine learning

Online veröffentlicht: 07. Okt. 2023

Seitenbereich: -

Eingereicht: 06. Jan. 2023

Akzeptiert: 17. Apr. 2023

DOI: https://doi.org/10.2478/amns.2023.2.00504

SchlüsselwörterConvolutional neural network, Machine learning, Maximum global pooling, Global averaging, Loss function, Image data processing

© 2023 Haifeng Mei et al., published by Sciendo

This work is licensed under the Creative Commons Attribution 4.0 International License.

Schlüsselwörter
Convolutional neural network, Machine learning, Maximum global pooling, Global averaging, Loss function, Image data processing