Applied Mechanics and Materials Vols. 668-669

Abstract: This article put forward a NCM_Apriori algorithm, which through compressing matrix and reducing the scan times to reduce the database I/O overhead, effectively improve the efficiency of association rule mining. At the same time in the process of generating association rules, computation is greatly reduced by using the nature of probability. And applies the algorithm to the mining of students' course selection system, which can provide decision support for colleges and universities.

Abstract: With the rapid development of transportation, more application of computer, communication and intelligent detection technology appeared in transportation field. The license plate recognition system based on template matching is researched in this paper. The preprocessing to the image such as binarization, tilt compensation and edge extraction has been finished, and then the positioning and segmentation of image are realized. After the normalization to the image, the license plate recognition result based on template matching method is obtained through experiments.

Abstract: Compressive sensing (CS) implements sampling and compression to sparse or compressible signals simultaneously. Compressive signal processing is a new signal processing scheme base on compressive sensing theory. In this paper, the problem of signal compressive reconstruction base on narrow-band interference is researched. The reconstruction performance of BP, MP, and OMP algorithms with narrow-band interference is analyzed by computer simulations.

Abstract: There are the results of the acoustic emission (AE) signals spectrum modification during steel 30XGSA heating near the austenite transformation intervals. It is established in the AE signals spectrum during heating depend on steel phase structure. It makes possible the application of AE method for determination the austenite transformation borders. The critical points of austenite formations and austenite decay are based on heat treatment technologies. It is known, that there are phenomenon of steels properties modification without structure modification not for from phase transformation borders. This was very convenient for detecting the particles with foregone energy characteristics.

Abstract: On the basis of the spline function theory and the study of spline functions, the article discusses the refinement shift-invariant space of the functions in detail. The important quality and characteristics of the space are acquired to offer the methods for wavelet construction.

Abstract: The paper studies the Cauchy problem for impulsive transport equations. The solution of the equations is obtained by using the iterative method and the characteristic method. An example is given to illustrate our result.

Abstract: A new recognition method based on Gaussian mixture model for speech emotion recognition is proposed in this paper. To improve the effectiveness of feature extraction and accuracy of emotion recognition, extraction of Mel frequency cepstrum coefficient combined with Gaussian mixture model is used to recognize speech emotion. According to feature parameters extraction method by analyzing the principle of vocalization theory, emotion models based on Gaussian mixture model are generated and the similarity of their templates is obtained. A series of experiments is performed with recorded speech based on Gaussian mixture model and indicates the system gains high performance and better robustness.

Abstract: In this paper, we employ semi-discrete finite element method to study the convergence of the Cauchy equation. The convergent order can reach. In numerical results, the space domain is discrete by Lagrange interpolation function with 9-point biquadrate element. The time domain is discrete by two difference schemes: Euler and Crank-Nicolson scheme. Numerical results show that the convergence of Crank-Nicolson scheme is better than that of Euler scheme.

Abstract: In this paper, we discuss the upper semi-continuity of the solution to parameter η-Set-valued weak vector variational inequality problem. We show that the operator of parameter η-Set-valued weak vector variational inequality is not continuous, but it satisfies ν-semicontinuous and η-weak C pseudo-monotone. Our results generalize the previous results in the literature.

Abstract: In order to automatically expand high quality polarity adjectives as much as possible by using the WordNet, three types of query strings are used in this study. Among them, the third type of query string is the most important. Besides ‘synonym’ and ‘antonym’ relations, we also use ‘similar to’ and ‘also see’ relations of the WordNet. In fact, the last two relations are more important than the first two ones. With the help of the newly proposed method, we can calculate the sentiment intensity of those expanded adjectives exactly. Abundant experiments demonstrate that our improved POAE algorithm is very effective in expanding and ranking adjective seed list in the sentiment analysis.