The multitude of hardware and software applications generate a lot of data and burden security solutions that must acquire informations from all these heterogenous systems. Adding the current dynamic and complex cyber threats in this context, make it clear that new security solutions are needed. In this paper we propose a wrapper feature selection approach that combines two machine learning algorithms with an improved version of the Binary Bat Algorithm. Tests on the NSL-KDD dataset empirically prove that our proposed method can reduce the number of features with almost 60% and obtains good results in terms of attack detection rate and false alarm rate, even for unknown attacks.
Today, the never-ending stream of security threats requires new security solutions capable to deal with large data volumes and high speed network connections in real-time. Intrusion Detection Systems are an omnipresent component of most security systems and may offer a viable answer. In this paper we propose a network anomaly IDS which merges the Support Vector Machines classifier with an improved version of the Bat Algorithm (BA). We use the Binary version of the Swarm Intelligence algorithm to construct a wrapper feature selection method and the standard version to elect the input parameters for SVM. Tests with the NSL-KDD dataset empirically prove our proposed model outperforms simple SVM or similar approaches based on PSO and BA, in terms of attack detection rate and false alarm rate generated after fewer number of iterations.
Search engines have become a de facto place to start information acquisition on the Internet. Sabotaging the quality of the results retrieved by search engines can lead users to doubt the search engine provider. Spam websites can serve as means of phishing. This paper shows a spam host detection approach that uses support vector machines(SVM) for classification. We create a parallel version of standard Particle Swarm Optimization(PSO) to determine free parameters of the SVM classifier and apply our proposed model to a content web spamming dataset, WEBSPAM-UK2011. Our implementation of the parallel PSO is constructed on a pool of threads and each thread executes tasks associated to a particle from the swarm. Experiments showed that our proposed model can achieve a higher accuracy than regular SVM and outperforms other classifiers (C4.5, Naive Bayes). Furthermore, parallel version of standard Particle Swam Optimization(PSO) can efficiently select parameters for SVM.
scite is a Brooklyn-based organization that helps researchers better discover and understand research articles through Smart Citations–citations that display the context of the citation and describe whether the article provides supporting or contrasting evidence. scite is used by students and researchers from around the world and is funded in part by the National Science Foundation and the National Institute on Drug Abuse of the National Institutes of Health.