Extraction of Topological Features from Communication Networks: Difference between revisions
Created page with "Title: Extraction of Topological Features from Communication Networks Research Question: Can self-organizing feature maps be used to classify different types of communication network topologies? Methodology: The researchers compared three types of communication network topologies: regular, random, and scale-free. They represented these topologies using adjacency matrices and their eigenvalues. A self-organizing feature map neural network was used to classify the differ..." |
No edit summary |
||
| Line 9: | Line 9: | ||
Implications: The results suggest that self-organizing feature map neural networks can be used to analyze and classify communication network topologies. This could have practical applications in network management, security, and optimization. The study also highlights the potential of using eigenvalues as input features for neural networks, as they were found to be invariant to the order of node labels. | Implications: The results suggest that self-organizing feature map neural networks can be used to analyze and classify communication network topologies. This could have practical applications in network management, security, and optimization. The study also highlights the potential of using eigenvalues as input features for neural networks, as they were found to be invariant to the order of node labels. | ||
Link to Article: https://arxiv.org/abs/ | Link to Article: https://arxiv.org/abs/0404042v2 | ||
Authors: | Authors: | ||
arXiv ID: | arXiv ID: 0404042v2 | ||
[[Category:Computer Science]] | [[Category:Computer Science]] | ||
Latest revision as of 15:50, 24 December 2023
Title: Extraction of Topological Features from Communication Networks
Research Question: Can self-organizing feature maps be used to classify different types of communication network topologies?
Methodology: The researchers compared three types of communication network topologies: regular, random, and scale-free. They represented these topologies using adjacency matrices and their eigenvalues. A self-organizing feature map neural network was used to classify the different topological patterns.
Results: The study found that self-organizing feature map neural networks can effectively classify different types of communication network topologies. The neural network was trained on a dataset of 900 topological patterns, and it accurately classified the patterns into their respective categories.
Implications: The results suggest that self-organizing feature map neural networks can be used to analyze and classify communication network topologies. This could have practical applications in network management, security, and optimization. The study also highlights the potential of using eigenvalues as input features for neural networks, as they were found to be invariant to the order of node labels.
Link to Article: https://arxiv.org/abs/0404042v2 Authors: arXiv ID: 0404042v2