Applied Mechanics and Materials Vols. 713-715

Abstract: Efficient data mining model design for a large database in the cloud computing environment is studied. For large databases efficiently mining problem, an efficient data mining model in the cloud computing environment based on improved manifold learning algorithms is proposed. The use of nonlinear manifold learning algorithms is able to reduce dimensionality of data vector feature in cloud computing environments, through characteristic extraction module to preprocess data, improved classical manifold learning algorithm is adopted to increase the distance between the data of sample spread intensive area and shorten the distance between the data of sample spread sparse area, prompting even overall distribution of sample database under cloud computing environment, so as to achieve accurate mining for efficient data in cloud computing environment. The experimental results show that the proposed method can accurately mine target data under cloud computing environments, with high efficiency and precision.

Abstract: Cloud storage as a popular application in cloud services, are developing from single cloud service provider towards multiple cloud service providers. As a consequence, users need to manage their files from different sources and share in multi-clouds. While the conventional access control models are no longer suitable in multi-clouds environment. Therefore, a multi-clouds access control model based on XACML, named MCACM, is proposed here to implement access control in multi-clouds environment. In this model, cross-clouds access control is enabled through extending standard XACML framework. At last, we implemented a prototype of MCACM, and performance evaluation results show that our scheme is efficient.

Abstract: Network congestion control problem is one of the most key problems in network study. This paper studies different versions of TCP source port algorithm such as Tahoe, Reno, NewReno, SACK and Vegas, and makes simulation research on linear network and dumbbell network environment algorithm by using NS2 network simulator, and finally it concludes that: when there is no competition in the network, it is better to choose Vegas algorithm; when there is competition, choose Reno, NewReno, or Sack algorithm is more advisable.

Abstract: In this paper, we describe some privacy threats in the Internet of Things and some research works on privacy protection. We present a new scheme base on cryptosystem to protect privacy in the Internet of Things. The scheme includes location privacy protection, data privacy homomorphism mechanism and information hiding technology, and secure multi-party computation on data privacy.

Abstract: With the shortage of energy and global climate warming, as well as the low-carbon economy and green computing coming, the energy consumption of cloud computing has become a critical issue, and even the economic benefits of cloud computing has been widely discussion. In view of the phenomenon of energy is wasted seriously in cloud computing, the energy optimization techniques in cloud computing platform have been studied and summarized in this paper. The concept, characteristics and development of cloud computing are introduced firstly. And then the existing energy consumption optimization approaches of cloud computing are studied deeply. The opening and closing techniques, dynamic voltage adjustment technology, virtual energy-saving technology and resource scheduling optimization technology are studied deeply. Finally, the contents are summarized and the future is looked forward.

Abstract: RED algorithm is a kind of implementation which is relatively simple, and better solutions to optimize the network congestion problem, but the RED algorithm of control for the network changes is relatively sensitive, So in practical applications may be to enhance the practical utility of RED algorithm by artificially setting control parameters. In this paper, analysis the effect of maximum drop probability on the queue and average queue based on the discrete TCP/RED feedback network model. And discuss when the queue length stabilized at the desired target, the relationship of the maximum drop probaility value with other network parameters. Based on theoretical analysis, proposed an improved RED algorithm, based on NS-2 simulation results show that this algorithm has good average latency and packet loss rate in a complex network environment, superior FRED and BLUE algorithm.

Abstract: Considering a multicast scenario, we want to minimize the coding links used for network coding while achieving the desired throughput. This article solves the optimization of the network coding to get with the lowest coding link scheme by using APGA. According to the simulation results, it can be concluded that the better performance of APGA than the previous algorithms.

Abstract: In the face of complex problems of implementation intentions predict the online shopping behavior, we conducted a preliminary exploration of group differences intention on the online shopping behavior using data mining method. The research results show, different types of temperament, personality, gender and living area in online shopping experience and behavior intention have common characteristics of groups, and different groups have obvious difference in online shopping behavior . Finally, this paper combined with large data and the mobile Internet era characteristic, put forward to large data comprehensive online shopping search behavior index, O2O business model as the foundation, the prospect of research on the construction of implementation intention theory of online shopping behavior.

Abstract: We take the overall architecture of internetware on-line evolution model as basic, and study on trust metric model of the software in internetware system. In view of the not accurate results from the rough and existing trust metric model granularity, this paper proposed a multi service and hierarchical dynamic trust metric model based on time frame. Model also offer a method to established time frame weighted factor based on inducing ordered weighted operator, which makes the trust measurement results more accurate. The trust measurement results obtained from the model will be used as decision-making basis for Bias game model.

Abstract: With the rapid development of modern computer technology, network has become people to pass information to each other a fast channel. The network information resources is very large, however, the people to find information resources needed to bring a lot of inconvenience. After recent years of rapid development of network information retrieval, on solving this problem have played a role, to a certain extent, meet the needs of the user to retrieve information, but in the test or not on the standard rate. This article through to the concept of data mining method, discusses the application of data mining in current network information retrieval. Data mining is analyzed with the combination of information filtering and visualization technology, thus improve the efficiency of network information retrieval, make up for the inadequacy of existing retrieval tools.