Applied Mechanics and Materials Vols. 513-517

Abstract: The problem of face classification is essentially a nonlinear subspace classification problem. The features of different face samples lie on different nonlinear subspaces. If the most representative samples of each subspace are selected as the training set, we can enhance the reliability of the classification, as well as reduce the computation. In this paper, a manifold based active learning algorithm, called Laplacian Transductive Optimal Design (LTOD), is presented to select the most representative samples. LTOD reconstructs each sample by the graph Laplacian matrix to make sure that the reconstructed samples and the original samples share the same manifold structure. Then, the samples which can be used to best reconstruct the whole sample set are selected as the training set. The experimental results on Yale face database have demonstrated the effectiveness of our method.

Abstract: The most weakness link in credible monitoring is that how to process multidimensional dynamic behavior data effectively. System behavior monitoring often needs to deal with different kinds of behavior data and those data can adopt status snapshot in multi-dimensional vector form to express. Obviously, data has strong useful knowledge information, which is regarded as a kind of classification ability. So we need to finish the mapping and classification between a variety of network behavior snapshot and dependable level. This paper introduces on network state snapshot owning the characteristics of high dimension, heterogeneous and dynamic and uses the theory of interval intuitionistic fuzzy to judge credible degree in the system and generate behavior quality trust level of nodes.

Abstract: Network computing system exists security issue, so we proposed a calculation method which could distinguish credible level from information mining. The calculation method uses system credible state detector to detect network system entities behaviors which in different environments and the different angles, extraction behaviors that affect credible decision-making as evidence. Normalized the systematic evidence, then the information of data mining based on cloud model and calculation credible identification. Finally, make the decision of system confidence level.

Abstract: Topic detection in news blogs needs to carefully arrange the information and analyze the characteristics of topics. However, there are some difficulties for hot topic detection in blogs. On one hand, information overload and dynamic change of web pages are obstacles of information arrangement. On the other hand, there are different hotness evaluation norms for web topics. The proposed method first analyzes the characteristics of the news blog and recognizes the factors which can influence the evolution of a topic. Then a word network is constructed, and candidate topics are extracted from the word network based on the complex networks theory. Finally, hot topics in the news blog are identified by measuring the user participation, opinion communication between users and user forgetting degree.

Abstract: To solve the problem of updating the software for Vehicle Fault Diagnose Instrument frequently, paper proposed an approach to realize the technology of parsing and matching vehicle diagnostic information based on XML database. Paper discussed the theory and process of the approach and used it to develop the Vehicle Diagnose Instrument based on Android OS. The testing result shows that this approach could be easily realized and improved the efficiency of updating the vehicle information.

Abstract: This paper analyzes the problems of the modern logistics: track and trace of products, and provides an effective solution that combines with RFID and web services to develop a B/S platform of track and trace of logistics . RFID middleware is adopted in the system implementation process. The paper has done a lot of research in some key technologies in the design ,the manufacturer, the third part of logistics and the retailers are the three key roles, they can both use this system to track and trace products in the Internet.

Abstract: The development and application of the Internet of things and the basic structure of wireless sensor network are briefly introduced,routing protocol in wireless sensor network is studied in detail. Since the wireless sensor network is a kind of energy-limited and resource-limited network, the routing protocol must guarantee the smaller routing expenses and the reducing energy consumption. Based on the analysis and research of existing wireless sensor network routing protocol, this paper emphasizes on the detailed analysis of the hierarchy routing protocol LEACH algorithm.The paper also introduces two improved algorithms according to the limit of LEACH protocol.

Abstract: Issues for food traceability are receiving enthusiastic research interests recently. Traceability system is aimed to provide information visibility through the farming, production, packing, distribution, transportation, and sales process. The objective of this research is to develop a mobile data collection system to enhance the efficiency of e-traceability data construction. The mobile system consists of a front-end application service with cell phone and QR code labels. By scanning the QR code labels related to the farming operations, traceability data can be uploaded simultaneously to the back-end web server. The miscellaneous traceability data construction can be simply replaced by scanning QR code labels. Applications of this mobile system can significantly reduce the complexity of traceability in agriculture.

Abstract: Augmented reality (AR) technology is one of the hot issues in the field of computer vision, There are many technologies for the development of online virtual display, Such as Flash, Java and Web3D technology. Web3D technology is an emerging network3D technology, in which VRML and Cult3D technologies are commonly used. This paper chooses a convenient interactive technologyCult3D, Cult3D Designer platform is given in a 3D virtual scene and product display of the implementation process, through a set of instructions to achieve user interactions with the product and the virtual museum roaming.

Abstract: The purpose of this article is to familiarize EFL teachers with the effectiveness of using the art of discourse with computer-assisted instruction (CAI) due to its advantages of multimedia, instantaneousness, interaction, flexibility and variety. CAI and the art of discourse are integrated to enhance the effect of college EFL teaching and learning.