Observatorio de Investigación y Desarrollo Tecnológico


  • ADVANCES IN INTELLIGENT SYSTEMS AND COMPUTING//(SPRINGER BOOK SERIES. AISC, VOLUME 1160)

    Volumen: 1160, Numero: 1, Páginas: 275-292 pp.

    MACHINE LEARNING AND DATA NETWORKS: PERSPECTIVES, FEASIBILITY, AND OPPORTUNITIES

    Abstract

    Currently, Machine Learning has become a research trend around the world and its application is being studied in most fields of human work where it is possible to take advantage of its potential. Current computer networks and distributed computing systems are key infrastructures that have allowed the development of efficient computing resources for Machine Learning. The benefits of Machine Learning mean that the data network itself can also use this promising technology. The aim of the study is to provide a comprehensive research guide on networking (networking) assisted by machine learning to help motivate researchers to develop new innovative algorithms, standards, and frameworks. This article focuses on the application of Machine Learning for Networks, a methodology that can stimulate the development of new network applications. The article presents the basic workflow for the application of Machine Learning technology in the field of networks. Then, there is also a selective inspection of recent representative advances with explanations of its benefits and its design principles. These advances are divided into several network design objectives and detailed information on how they perform at each step of the Machine Learning Network workflow is presented. Finally, the new opportunities presented by the application of Machine Learning in the design of networks and collaborative construction of this new interdisciplinary field are pointed out.


    Keywords


    Machine learning, Learning model, Traffic prediction, Traffic classification, Congestion control


    Resumen

    No disponible


    Palabras Clave


    No disponible





    Acceder