A Hybrid Approach Towards Intrusion Detection Based on Artificial Immune System and Soft Computing
This work addresses intrusion detection for network security, but it appears incremental as it builds on existing AIS and Soft Computing methods without claiming major breakthroughs.
The paper tackled the problem of intrusion detection by combining Artificial Immune System (AIS) and Soft Computing approaches to leverage adaptability and computational power, resulting in a system designed to effectively detect intrusions in networks.
A number of works in the field of intrusion detection have been based on Artificial Immune System and Soft Computing. Artificial Immune System based approaches attempt to leverage the adaptability, error tolerance, self- monitoring and distributed nature of Human Immune Systems. Whereas Soft Computing based approaches are instrumental in developing fuzzy rule based systems for detecting intrusions. They are computationally intensive and apply machine learning (both supervised and unsupervised) techniques to detect intrusions in a given system. A combination of these two approaches could provide significant advantages for intrusion detection. In this paper we attempt to leverage the adaptability of Artificial Immune System and the computation intensive nature of Soft Computing to develop a system that can effectively detect intrusions in a given network.