A Deep Learning Approach to protect critical infrastructures in developing countries: case of Senegalese public services sector

Over the past decade, developing countries, and Senegal in particular, have been faced with an upsurge in cyber-attacks on their critical infrastructure systems, especially in the public services sector. Given the lack of appropriate cyber defence mechanisms and the central role played by this sector in society, the challenges of protecting it against cyber-attacks, which have become more complex and sophisticated, are bound to multiply. This article proposes a new model for cyber defence of the public services sector in developing countries, particularly Senegal. The approach adopted is based on Deep Learning integrated into Wazuh, an open source SIEM. The proposed system was able to detect and neutralise cyber threats with an average accuracy of 97,40%. The dataset chosen for this study is a LOG dataset made up of log events collected from a KAGGLE dataset database and GitHub repositories.