@article{scholars10741,
           title = {Dynamic Ant Colony System with Three Level Update Feature Selection for Intrusion Detection},
             doi = {10.6633/IJNS.201801.20(1).20},
            note = {cited By 21},
          volume = {20},
          number = {1},
           pages = {184--192},
       publisher = {Femto Technique Co., Ltd.},
         journal = {International Journal of Network Security},
            year = {2018},
             url = {https://www.scopus.com/inward/record.uri?eid=2-s2.0-85057079811&doi=10.6633\%2fIJNS.201801.20\%281\%29.20&partnerID=40&md5=64c085ebfab8ac1cf134b6cbc755a8a3},
        keywords = {Ant colony optimization; Classification (of information); Computer crime; Intrusion detection; Network security, Anomaly based detection; Anomaly-based detections; Classification accuracy; Classification algorithm; Digital information; Feature selection algorithm; Intrusion Detection Systems; Network technologies, Feature extraction},
        abstract = {The current era is known as the age of digital information and general medium of access to this information is computer networks. The uses of network technology also make information insecure. Intrusion Detection System (IDS) has been proven effective against such attacks. The anomaly-based detection method is good to detect new attacks. One of the foremost shortcomings in the anomalybased detection is the irrelevant and redundant features to the classification algorithm that results in low detection rate. Therefore, the primary objective of the feature selection process is to enhance the classification accuracy by removing redundant and irrelevant features. In this research a new feature selection algorithm called, Dynamic Ant Colony System with Three Level Update Feature Selection, has been proposed. The proposed method uses a different level of pheromones that help ants to find the robust features. The method also utilizes the information of each individual ant during feature selection process and incorporates the accuracy of the classification algorithms. Results showed that proposed feature selection algorithm outperformed compared to the previous feature selection algorithms. {\^A}{\copyright} The Japan Society for Analytical Chemistry.},
          author = {Rais, H. M. and Mehmood, T.},
            issn = {1816353X}
}