A Survey on Routing Algorithms Based on Machine Learning

Abstract

With the rapid development of the Internet in today's world, many computer applications have emerged, including many new network applications, such as real-time multimedia streaming services. However, traditional computer routing algorithms have defects in these emerging network applications, so we urgently need to develop new routing algorithms to compensate for them. Machine learning technology has achieved considerable results in computer vision, image generation, and game processing. In order to further improve network performance, some research groups have introduced machine learning techniques into routing problems. Unlike traditional routing algorithms, machine learning-based routing algorithms are typically driven by data. Hence, they are more adaptable to changes in network status and can make decisions promptly. Current intelligent routing algorithms demonstrate their potential through the application of machine learning, and this idea is likely to become an indispensable part of the Internet in the future. This article first introduces machine learning technology, then introduces routing algorithms using different machine learning technologies, and finally points out the development prospects of routing algorithms based on machine learning technology.