Machine Learning Driven Network Routing

The research community have proposed some frameworks and use-cases for using machine learning in networking. However, there are still real-time problem and ML model selection problem. In this paper, we present the basic model of machine learning driven network routing. This model divided route optimization into the Optimization of Routing Protocol Parameter(ORPP) and the Optimization of Routing Efficiency and Quality(OREQ). The input and output of the machine learning model for routing optimization problems can be described as traffic matrix and route matrix. In the end, we present two experimental results of ORPP and OREQ to demonstrate this model's feasibility.