Open Access

Construction of data security protection model in archive informatization system based on deep learning

Min Feng

Min Feng

Archives Sichuan Normal UniversityChengdu, China
Search for this author on
Sciendo|Google Scholar
Feng, Min
  
Mar 21, 2025
Construction of data security protection model in archive informatization system based on deep learning's Cover Image
About this article
Previous Article
Next Article
Cite
Download Cover
Published Online: Mar 21, 2025
Received: Nov 06, 2024
Accepted: Feb 09, 2025
DOI: https://doi.org/10.2478/amns-2025-0597
Keywords
© 2025 Min Feng, published by Sciendo
This work is licensed under the Creative Commons Attribution-NonCommercial-NoDerivatives 4.0 International License.

Figure 1.

The single use of households takes time to change the number of users
The single use of households takes time to change the number of users

Figure 2.

The change of the number of individual gradient and the time consuming
The change of the number of individual gradient and the time consuming

Figure 3.

The single use of households takes time to change the rate of lag
The single use of households takes time to change the rate of lag

Figure 4.

The number of single accounts is said to change the number of users
The number of single accounts is said to change the number of users

Figure 5.

Changes in the growth of the average gradient and the amount of data transfer
Changes in the growth of the average gradient and the amount of data transfer

Figure 6.

The number of single accounts is said to change the rate of drop
The number of single accounts is said to change the rate of drop

Figure 7.

Comparison of PDEC and DEC-C on different data sets
Comparison of PDEC and DEC-C on different data sets

Figure 8.

The convergence velocity of different data sets is compared
The convergence velocity of different data sets is compared

The characteristics of several methods compare the results

Characteristic PPNPC CPFED FEDOPT PDEC
Low communication cost + + +
Privacy protection + + + +
Complicity + + +
Apply the IID data +
Device drop + + +
Fast convergence + + +

Communication overhead of cloud servers

|G|=1000,R=0.1 |U|=100,R=0.1 |U|=100,|G|=1000
Number of users Total transmission data(×108Byte) Single-use household gradient Total transmission data(×108Byte) Drop rate Total transmission data(×107Byte)
Server communication overhead 100 0.707 1000 0.738 0.00 7.812
150 1.089 1500 1.100 0.05 7.532
200 1.422 2000 1.463 0.10 7.275
250 1.805 2500 1.785 0.15 6.995
300 2.167 3000 2.147 0.20 6.739
350 2.510 3500 2.510 0.25 6.459
400 2.872 4000 2.882 0.30 6.217
450 3.255 4500 3.265 0.35 5.938
500 3.597 5000 3.597 0.40 5.674

Calculation overhead of cloud server

|G|=1000,R=0.1 |U|=100,R=0.1 |U|=100,|G|=1000
Number of users Overall running time(×105ms) Single-use household gradient Overall running time(×105ms) Drop rate Overall running time(×104ms)
Server overhead calculation 100 0.501 1000 0.498 0.00 5.033
150 0.712 1500 0.738 0.05 4.731
200 0.900 2000 0.958 0.10 4.459
250 1.134 2500 1.191 0.15 4.218
300 1.346 3000 1.461 0.20 4.021
350 1.565 3500 1.638 0.25 3.765
400 1.822 4000 1.883 0.30 3.507
450 2.011 4500 2.116 0.35 3.221
500 2.222 5000 2.400 0.40 2.995
Language:
English
Publication timeframe:
1 times per year
Journal Subjects:
Life Sciences, Life Sciences, other, Mathematics, Applied Mathematics, General Mathematics, Physics, Physics, other
Journal RSS Feed