Quality-of-Service Performance Comparison: Machine Learning Regression and Classification-Based Predictive Routing Algorithm
N.A.M. Radzib M.A.Ridwana · Turkish Journal of Computer and Mathematics Education (TURCOMAT) · 2021
The Internet network has evolved rapidly and by no means of slowing down. The complexity of the network isexpected to grow exponentially, and with the high dependencies on Internet applications, there is a need to upgrade the currentrouting mechanism in the network. The conventional routing protocol that is based on the shortest path is no longer relevant.Recently, Machine Learning (ML) algorithms have become more prevalent in networking due to their ability to solve complexproblems intelligently. This work proposes two ML predictive routing algorithms using regression and classificationapproaches to improve the Quality of Service in the network. Our simulation results show that the proposed regression-basedrouting improves the delay by up to 52% compared to the classification-based algorithm. Although the regression-basedrouting achieved better performance compared to the classification approach, it requires more input features to be trained. Thiswork also discusses the pros and cons of both approaches.