graph clustering

Today, applications of the virtualization technology are rapidly growing in which setting up and running multiple operating systems on a single physical system. Computational clouds are the most hallmark of this technology. Intrusion detection systems play a key role in protecting cloud resources on virtual machines. Regarding the increasing speed and complexity of these machines, it is necessary to increase the ability and accuracy of intrusion detection systems to identify different types of attacks at the right time. In this regard, the use of behavior-based approaches has attracted more attention due to their high scalability in large networks. The methods for intrusion detection that utilizes network traffic graph clustering do not have the accuracy and appropriateness with the speed of data transfer in the current computer networks. Thus, the solutions can be improved by choosing an appropriate strategy for clustering. In this paper, a new behavior-based method for detecting intrusion in computer networks is presented. To this end, the network data was modeled through the flow of data as a traffic distribution graph and then clustered using an improved Markov-based algorithm. Then, the produced clusters are used to construct an intrusion detection model by analyzing a set of modified statistical criteria. The proposed model was examined and evaluated on the DARPA 99 dataset and compared with seven other robust methods. The results show that the proposed method detects attacks with high accuracy and works better than the methods which do not use the graph clustering.

Due to the growing desire of people to join and use social networks, communication and shar ing data in these networks has been considered by various sciences such as political science, psychology, sociology, economics, etc. Hence, researchers have begun to distinguish and extract relationships between individuals from the data contained in these networks, to create more accurate communities. However, there is still no effective method to identify and extract communities based on social media data. In this article, a method has been proposed for social network accurate clustering by using Imperial Competitive Evolutionary Algorithm (ICEA) and selecting the initial population based on the density-based clustering criterion. The proposed method has improved the result of modularity about 21.45% in average, compared to rival basic ICEA and extracted more densed communities.

In recent years, power system islanding is considered as the last resort to prevent network instability. The purpose of controlled islanding is to establish self-sustaining electrical islands to prevent a blackout and lead to ease of restoration. In this regard, an appropriate decision - making algorithm is required to determine the best separation points in a short time.
This paper proposes a novel methodology by developing the hierarchical spectral clustering algorithm. This method is computationally very efficient and determines an islanding solution with minimal power flow disruption while ensuring that each island contains only coherent generators. The proposed method also enables operators to constrain any branch to be excluded from the islanding solution. The method is tested using the models of the IEEE 39 and IEEE 118-bus test systems and comparative analysis with existing methods have been performed. The simulation results demonstrate that the methodology is more efficient than another existing approach and could be utilized in the real-time application of the power grid.