Cyber-Enabled Distributed Computing and Knowledge Discovery 

Abstract: Data clustering is a frequently used technique in finance, computer science, and engineering. In most of the applications, cluster sizes are either constrained to particular values or available as prior knowledge. Unfortunately, traditional clustering methods cannot impose constrains on cluster sizes. In this paper, we propose some vital modifications to the standard k-means algorithm such that it can incorporate size constraints for each cluster separately. The modified k-means algorithm can be used to obtain clusters in preferred sizes. A potential application would be obtaining clusters with equal cluster size. Moreover, the modified algorithm makes use of prior knowledge of the given data set for selectively initializing the cluster centroids which helps escaping from local minima. Simulation results on multidimensional data demonstrate that the k-means algorithm with the proposed modifications can fulfill cluster size constraints and lead to more accurate and robust results.

Abstract: Cloud computing is a rapidly growing internet technology for facilitating with various services to its consumers. It proposes many striking promises to the general public or big companies like Amazon, Google, Microsoft, and IBM etc. to maintain and upgrade their position in fast-growing cloud computing environment and to enhance their services for a large number of users. However, with the fast development and enticing offerings, many issues associated with this technology also arises which needs to be addressed with security being the strongest barrier to its adoption. Security concerns are an active area of research, which needs to be addressed properly to avoid security threats and attacks which are disasters for both service providers and service consumers. This paper highlights cloud computing architectural principles, cloud computing key security requirements, cloud computing security threats and cloud computing security attacks with their mitigation techniques, and future research challenges.

Abstract: Cloud computing has attracted great interest from both academic and industrial communities. However, only a few efforts have been devoted to building tools for supporting Cloud service discovery. In this paper, we present a four-stage, agent-based Cloud service discovery protocol. Utilizing an ontology description, in which each resource is described semantically and relatively to other resources, we develop a multi-agent system that cooperates efficiently by introducing a flexible ontology-based matching. To increase the utility and success rate of matching consumers' requests to resources, we use a database to store and keep track of historical data for making intelligent recommendation based on attribute value prediction. Empirical results show that when broker agents in our system use a Cloud ontology and a connection procedure with a recommendation stage achieved better performance in finding the appropriate Cloud services than when broker agents 1) use only the connection procedure but not the Cloud ontology and 2) do not use both the connection procedure and the Cloud ontology.

Abstract: The Internet of Things (IoT) has been considered as the third revolution in the digital technology after the computer and the Internet, and it brings significant advantages to smart grid. Disaster prediction and prevention of power-lines is one of the most difficult problems for electricity transmission companies. Advanced sensing and communication technologies of IoT can effectively avoid or reduce the damage of natural disasters to the power-lines, and hence improve the reliability and stability of power transmission. In this paper, a three-layer architecture model of the Internet of things for smart grid (SG-IoT) is proposed. An overview of the key technologies in SG-IoT is given. An IoT based power-lines online monitoring system is presented.