.png){kind=icon}
.png){kind=icon}
Return
Comparison of Machine Learning Algorithms for SDN Optimization Using TOPSIS Methodology
.png){kind=icon}
PRE
AI
DOI：10.1007/978-3-031-98768-7_22.png)
.png){kind=icon}
Original request for help.png){kind=icon}
Share.png){kind=icon}
Save
Abstract
En 中文
The article covers how machine learning using the TOPSIS methodology can help us select the most appropriate algorithm for optimizing software-fined networks, highlighting machine learning in combination with the TOPSIS methodology as a valuable tool to facilitate this selection. Soft-ware-defined networks have advanced significantly in how they are managed. Their automation and efficiency make them ideal for optimization through machine learning, thus optimizing repetitive and complex tasks and leaving the network administrator free to focus on more strategic activities. Manual optimization of software-defined networks is inefficient and prone to errors and high operation and maintenance costs, so machine learning provides automated solutions, and the TOPSIS methodology will help us select the most appropriate algorithm for optimizing software-defined networks. TOPSIS has multiple solutions, one of them visual tools to facilitate the comparison of algorithms through a graph of solutions that allows one to identify the best and worst algorithms. Despite its significant advantages, this TOPSIS methodology can be complex to interpret and costly when there are problems with many alternatives. This proposed approach, which focuses on machine learning and network optimization and uses TOPSIS methodology, is positioned as a critical strategy to analyze and solve the inefficiency of manual management of devices in software-defined networks, thus improving the network's performance, security, and confidentiality.
Keywords:
TOPSIS
Automation
Efficiency
Performance
Security
Journal
P