Data-driven Intrusion Detection System for Small and Medium Enterprises
Journal article, Peer reviewed
Accepted version
Åpne
Permanent lenke
http://hdl.handle.net/11250/2627244Utgivelsesdato
2019Metadata
Vis full innførselSamlinger
Originalversjon
10.1109/CAMAD.2019.8858166Sammendrag
Small and Medium Enterprises (SMEs) have become targets of attack by cyber criminals in resent times. This paper therefore aim to address awareness and challenges of SMEs related to IDSs as the most important defense tool against sophisticated and ever-growing network attacks. An IDSs framework was actually introduced for efficient network anomaly detection for SMEs and provided experimental results to illustrate the benefits of the proposed framework. The proposed framework deals with one of the main challenges that IDSs of SMEs are facing, the lack of scalability and autonomic self-adaptation. Training, testing and evaluation of IDSs applying different machine learning (ML) techniques are presented. Results of experiments show that using feature selection approaches can lead to better classification accuracy and improved computational speed.