Data-driven Network Connectivity Analysis: An Underestimated Metric
Uloženo v:
| Název: | Data-driven Network Connectivity Analysis: An Underestimated Metric |
|---|---|
| Autoři: | Xu, J, Nair, DJ, Jayakumar Nair, Divya |
| Zdroj: | IEEE Access, Vol 12, Pp 60908-60927 (2024) |
| Informace o vydavateli: | Research Square Platform LLC, 2024. |
| Rok vydání: | 2024 |
| Témata: | large-scale data, relative size of largest connected component (RSLCC), network structure analysis, anzsrc-for: 4605 Data Management and Data Science, 05 social sciences, anzsrc-for: 46 Information and Computing Sciences, 0211 other engineering and technologies, 02 engineering and technology, TK1-9971, 4605 Data Management and Data Science, anzsrc-for: 40 Engineering, Network connectivity metric, deep neural network (DNN), 46 Information and Computing Sciences, 0502 economics and business, network disruption degree (NDD), Generic health relevance, Electrical engineering. Electronics. Nuclear engineering, anzsrc-for: 09 Engineering, anzsrc-for: 08 Information and Computing Sciences, anzsrc-for: 10 Technology |
| Popis: | In network structure analysis, metrics such as Isolated Node Ratio (INR), Network Efficiency (NE), Network Clustering Coefficient (NCC), Betweenness Centrality (BC), and Closeness Centrality (CC) are used as quantitative tools to measure network connectivity. However, there is another metric that is often easily overlooked and underestimated, i.e., the Relative Size of Largest Connected Component (RSLCC), we do not find any literature that analyzed RSLCC in a separate study. However, through the research in this paper, we not only prove that this metric is underestimated, but also design 7 methods to predict the value of this metric, with a Deep Neural Network (DNN) prediction accuracy of more than 99%. This research results can be applied to any network, and in a disaster scenario, whether it is a physical entity network or a virtual abstract network, the approximate network connectivity value can be predicted simply by knowing the number of connected edges in the pre-disaster network and the number of connected edges in the post-disaster network. |
| Druh dokumentu: | Article |
| Popis souboru: | application/pdf |
| ISSN: | 2169-3536 |
| DOI: | 10.21203/rs.3.rs-3978886/v1 |
| DOI: | 10.21203/rs.3.rs-3978886/v2 |
| DOI: | 10.1109/access.2024.3393968 |
| Přístupová URL adresa: | https://doaj.org/article/e62b4d8ca7df4847814a393e980cf4d5 |
| Rights: | CC BY CC BY NC ND |
| Přístupové číslo: | edsair.doi.dedup.....6024f85bb1cf9696f6ccf8ea06286b1a |
| Databáze: | OpenAIRE |
Full Text Finder 
Nájsť tento článok vo Web of Science | Abstrakt: | In network structure analysis, metrics such as Isolated Node Ratio (INR), Network Efficiency (NE), Network Clustering Coefficient (NCC), Betweenness Centrality (BC), and Closeness Centrality (CC) are used as quantitative tools to measure network connectivity. However, there is another metric that is often easily overlooked and underestimated, i.e., the Relative Size of Largest Connected Component (RSLCC), we do not find any literature that analyzed RSLCC in a separate study. However, through the research in this paper, we not only prove that this metric is underestimated, but also design 7 methods to predict the value of this metric, with a Deep Neural Network (DNN) prediction accuracy of more than 99%. This research results can be applied to any network, and in a disaster scenario, whether it is a physical entity network or a virtual abstract network, the approximate network connectivity value can be predicted simply by knowing the number of connected edges in the pre-disaster network and the number of connected edges in the post-disaster network. |
|---|---|
| ISSN: | 21693536 |
| DOI: | 10.21203/rs.3.rs-3978886/v1 |