The authors use Passive DNS to log domain name query history of real campus network environment, and construct eighteen feature sets grouping by diversity, time, growth, and relevance, and then propose a model detect Fast-Flux Domains using random forest algorithm. The result shows that the proposed model can classify domains with accuracy over 90% by cross validation experiments. The model can detect Fast-Flux domains in the datasets used in this study more effectively compared with Fluxbuster.

This paper briefly analyzes the characteristics and evolution of the Science and Technology Parks in developed countries and in China, irrespectively. Compared with the former, China's state-level Science and Technology Parks have more complex context and more problems in development process. China's high-tech industry has grown quantitatively in various aspects.However, the industry as a whole is still at the low-end of the global value chain. With an innovation cluster perspective, the paper proposes the ideal development path of the Science and Technology Parks of China. The implication of innovation cluster, the importance of institution for collaboration and the influence of global driven force are explained.

The stability of basic leafset protocol is analyzed, which is a P2P overlay maintenance protocol. By analyzing the protocol behavior, it proves that the overlay topology must reach steady states after running the protocol for finite time and the final topology is one of the three types: ring, loopy and partitioned. This conclusion is also in accordance with the experimental results of basic leafset protocol.

Because of the inherent node (link) failures in wireless networks, virtual backbones should be fault-tolerant. Fault-tolerant virtual backbones were modeled as 2-connected k-dominating sets. An approximation algorithm was designed to find a 2-connected k-dominating virtual backbone in wireless ad-hoc networks by analyzing the properties of maximal independent sets in unit disk graphs and the block-cutvertex tree structure of connected graphs. The time complexity and the performance ratio of the algorithm were analyzed and proved to be a constant, respectively.