Applied Mechanics and Materials Vols. 713-715

Abstract: We consider how to efficiently text classification on all pairs of documents. This information can be used to information retrieval, digital library, information filtering, and search engine, among others. This paper describes text classification model which based on KNN algorithm. The text feature extraction algorithm, TF-IDF, can loss related information between text features, an improved ITF-IDF algorithm has been presented in order to overcome it. Our experiments show that our algorithm is better than others.

Abstract: Bifurcation which may lead to chaos is the typical character of nonlinear system, and an asymmetric system with asymmetric parameter is adopted in this paper. The basic characteristics which vary with the asymmetric parameter are investigated firstly, and then, the second-order averaging method is used to investigate the local bifurcation of the asymmetric system. The super and sub critical saddle-node bifurcation curves of both left center and right center of the system are solved analytically. The results show that the degree of asymmetric is influenced by the value of asymmetric parameter and the two bifurcation curves of the same center are intersected at the point which also depends on the asymmetric parameter value.

Abstract: Turn of the century, with the rapid development of database and network technology, we produce and the ability to collect data has been rapidly increasing, a large amount of data stored in databases and data warehouses, we have been submerged in the vast ocean of data and information. People need to have a new, more effective means to a variety of large amounts of data mining in order to realize their potential, data mining is to generate and rapidly developed applications in such environments, it appears to automatically and intelligently put massive data transformation provides a means for useful information and knowledge. This article will combining data mining and e-commerce, introduced the data in e-commerce information platform mining methods, so as to provide reference for the information network of employees, in order to better analyze the hidden relationships between data and model, master user preferences, provide personalized marketing decisions for decision support network platform, to reduce the risk.

Abstract: Digital watermarking technology has become an important means of integrity and respect for people's authenticity, as well as users of copyright and intellectual property security and other interests such as the protection of digital works. In this paper, we proposed a watermarking algorithm based on discrete wavelet transform (DWT) of image and singular valued composition (SVD). The original image is decomposed with DWT,the watermarking image is decomposed with SVD after chaotic scrambling,and then the singular values of watermarking are embedded into some coefficients of decomposed original image. In this algorithm, after decomposing the original host image into four bands, we apply the SVD to watermark image,and modify DWT coefficients of the host image with the singular values of the watermark image. The outstanding features of the proposed algorithm are that it provides larger watermarking capacity and is more robust than others.

Abstract: Gesture has different speed in the process of movement. To reflect the different speed of the user, this paper presents a gesture speed estimation method. Firstly, we use data glove and camera to establish the relation between the variation of gesture contour and that of gesture speed. Secondly, we build the gesture speed estimation model by stages. Finally, we get the real-time speed of hand motion through this model and complete interactive task. The main innovation of this paper is that we reveal the relation between the gesture contour and speed to lay the foundation for further capture the interaction of user intention. Experimental results indicate that the time cost of our method decreased by 31% compared with freehand tracking based on behavioral models and the 3D interactive system based on our model is of high the user experience.

Abstract: In view of the SIFT algorithm in image matching will produce a lot of mismatches, the paper has applied a method which is based on Hough Transform will remove the SIFT matching error effectively. Firstly, to use the SIFT algorithm finish the image matching roughly. And then, using the Hough Transform to form the equal division hough units. And according to the matching parameter to distribute all the match into the hough units. The match in the units which has least matching-pair will be deleted. Experimental results show that the method can effectively improve the matching accuracy of feature matching and it lays a foundation for the following robot vision navigation.

Abstract: In order to improve the modeling efficiency of RBF neural network, an Artificial Fish Swarm Algorithm (AFSA) training algorithm with an adaptive mechanism is proposed. In the training algorithm, the search step size and visible domain of AFSA algorithm can be adjusted dynamically according to the convergence characteristics of artificial fish swarm, and then the improved AFSA algorithm is used to optimize the parameters of RBF neural network. The example shows that, the proposed model is a better approximation performance for the nonlinear function.

Abstract: Tetrolets are Haar-type wavelets whose supports are tetrominoes which are shapes made by connecting four equal-sized squares. Firstly, the magnitude of polarization image and angle of polarization image can be decomposed into low-frequency coefficients and high-frequency coefficients with multi-scales and multi-directions using tetrolet transform. For the low-frequency coefficients, the average fusion method is used. For the each directional high frequency sub-band coefficients, the larger value of region energy information measurement is used to select the better coefficients for fusion. At last the fused image can be obtained by utilizing inverse transform for fused tetrolet coefficients. Experimental results show that the proposed method can detect image features more effectively and the fused image has better subjective visual effect.

Abstract: In the process of data mining, how to operate the data mining as well as protect the private data is a problem must be solved. This paper proposed an improvement of decision tree classification algorithm. Homomorphism encryption system, digital envelopes technology and secret sorting are applied to protect the data privacy. Our contribution is a privacy preserving protocol consist of homomorphism encryption system and secret sorting. Analysis shows that this algorithm can get right results as well as protect the privacy of the private data.

Abstract: Efficient resource scheduling in dynamic environment reveals several challenges due to its high heterogeneity, dynamic behavior, and space shared utilization. In order to address this issue, a scheduling model based on computational economy was proposed in this paper. Firstly, the problem of resource scheduling was analyzed, and the essence of resource scheduling was concluded. Secondly, a scheduling strategy based on computational economy was presented. It applied the principles of economics and broker technology in its resource scheduling process. The strategy of the model synthetically considered two factors: execution cost and total complete time. An evaluation function driven by user’s needs was also built based on the two factors. Finally, the scheduling strategy was simulated. The result shows that the model can adjust the relation between price and time based on computational economy efficiently. It indicates that the model is an effective approach for resource scheduling.