AI-Driven Cloud Honeypot Network

Abstract

This project introduced the cloud based honeypot system development and uses the data collected by the honeypot to create comprehensive machine learning models and understand patterns and fingerprints to eventually categorize this data into a potential threat or a malicious activity. By utilizing cloud services, the system can be scaled up and down as required, the architecture integrates advanced data capturing techniques and date pipelining for proper isolation of the data collected and also the machine learning models, the architecture is comprised of capturing services, machine learning service, dashboard and a data pipeline to help isolate the individual components and services, data captured through the capturing services is uploaded to the machine learning service for data classification and model making. The research addresses the critical issue of lack of data for machine learning models in security and attack patterns including cloud environments.