A Framework for near-Real-Time Intrusion Detection in Micro-Enterprise Environments

This paper presents a novel methodology for near-real-time intrusion detection tailored to micro-enterprise environments, where resources are limited but cyber risks are escalating. The experimental design uses the lightweight machine learning techniques in Weka with robust cross-validation to ensure reliable detection while minimising overfitting and underfitting. To complement this, we introduce the Agile Cybersecurity Maturity Model (ACMM), which supports adaptive and proactive security practices. The key contributions of this paper are: (i) a practical intrusion detection framework optimised for micro-enterprise constraints, (ii) the integration of machine learning and validation techniques for near-real-time monitoring, and (iii) the proposal of ACMM as a maturity model to strengthen long-term resilience. Together, these advances provide micro-enterprises with an affordable, scalable, and accessible defence against increasingly sophisticated threats, while supporting the protection of broader community and critical infrastructure.