The search for anomalies in network traffic

Nowadays, Internet networks are an integral part of our life and business. For Internet providers, a significant portion of profits depends on the level of customer satisfaction. Therefore, monitoring and controlling network anomalies plays a key role in improving the quality of the user experience. This paper provides a comparative analysis of creating a machine learning model using classification and clustering algorithms to analyze anomalies in network traffic. In this paper, algorithms such as gradient boosting and Density-Based Spatial Clustering of Applications with Noise (DBSCAN) were used. The efficiency of these algorithms in terms of working with Internet network data will also be described. Furthermore, the final solution will be presented as two different models trained using Jupyter Notebook and the low-code Artificial intelligence (AI) application platform of "AiB" company (hereinafter referred to as Razum AI). A comparative analysis of the development speed using these platforms, as well as the effectiveness of the resulting models, will also be presented. Thus, the second objective of the paper is to compare the aforementioned machine learning platforms.