DATA ANALYSIS AND MACHINE LEARNING IN CLOUD AND FOG PLATFORMS FOR EFFICIENT DATA TRANSMISSION

Abstract

This article investigates the effectiveness of traditional methods for optimizing data transmission and the proposed model based on machine learning algorithms. In the context of rapid and constant growth in the volume of information, expansion of network infrastructure and increasing requirements for system performance, there is a need for an adaptive approach to traffic management and reducing energy consumption. Special attention is paid to the integration of machine learning technologies in order to increase the efficiency of data transmission in cloud and fog computing. The proposed model combines network load prediction methods (LSTM, XGBoost), data compression technologies (VAE, K-Means) and software-defined networks (SDN) for traffic balancing. As part of the study, combinations of algorithms were tested in distributed environments with different load levels, which made it possible to evaluate the effectiveness of each according to the criteria of performance, reliability and energy consumption. This approach provides adaptability to dynamic changes in traffic and helps reduce data transmission delays. The obtained research results confirm the feasibility of machine learning in the application to the optimization of network interaction and open up prospects for further research in the field of intelligent control of fog computing. The proposed solution can be used for various needs, including autonomous transportation systems, medical information platforms, industrial Internet of Things and critical infrastructures.