These days, with the tremendous growth of network-based service and shared information on networks, the risk of network attacks and intrusions increases too, therefore network security and protecting the network is getting more significance than before. Intrusion Detection System (IDS) is one of the solutions to detect attacks and anomalies in the network. The ever rising new intrusion or attack types causes difficulties for their detection, therefore Data mining techniques has been widely applied in network intrusion detection systems for extracting useful knowledge from large number of network data to detect intrusions. Many clustering and classification algorithms are used in IDS, therefore improving the functionality of these algorithms will improve IDS performance. This paper focuses on improving KNN classifier in existing intrusion detection task which combines K-MEANS clustering and KNN classification.
Wireless sensor networks (WSNs) play a prominent role in the world of computer networks. WSNs rely on deployment as a basic requirement and an effective factor on the basic network services. In deployment, creating a balance between conflicting optimisation factors, e.g. connectivity and coverage, is a challenging and sophisticated issue, so that deployment turns into an NP-complete problem. The majority of existing researches has attempted to tackle this problem by applying classic single-objective metaheuristic algorithms in 2D small-scale uniform environments. In this study, a new hybrid multi-objective optimisation algorithm, which is constructed by the combination of multi-objective bee algorithms and Levy flight (LF) random walk is proposed to deal with the deployment problem in WSNs. For this purpose, two of the most important criteria, connectivity and coverage, have been considered as objectives. A series of experiments are carried out in large-scale non-uniform 3D environments, despite the fact that most of the present methods are applicable in small-scale uniform 2D environments. This study completely takes into account the stochastic behaviour of swarms, something that other papers do not consider. The evaluation results show that the multi-objective LF bee algorithm, in most cases, surpasses NSGAII, IBEA and SPEA2 algorithms.
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.