Pan H.,Xian University of Finance and Economics |
Yuan Y.,University of Houston
Statistics in Medicine | Year: 2016
The Bayesian model averaging continual reassessment method (CRM) is a Bayesian dose-finding design. It improves the robustness and overall performance of the continual reassessment method (CRM) by specifying multiple skeletons (or models) and then using Bayesian model averaging to automatically favor the best-fitting model for better decision making. Specifying multiple skeletons, however, can be challenging for practitioners. In this paper, we propose a default way to specify skeletons for the Bayesian model averaging CRM. We show that skeletons that appear rather different may actually lead to equivalent models. Motivated by this, we define a nonequivalence measure to index the difference among skeletons. Using this measure, we extend the model calibration method of Lee and Cheung (2009) to choose the optimal skeletons that maximize the average percentage of correct selection of the maximum tolerated dose and ensure sufficient nonequivalence among the skeletons. Our simulation study shows that the proposed method has desirable operating characteristics. We provide software to implement the proposed method. © 2016 John Wiley & Sons, Ltd.
Ma X.,Xian University of Finance and Economics
Journal of Theoretical and Applied Information Technology | Year: 2012
This paper analyzes the relationship between user's browsing behavior and interest. User interest is reflected by typical user's browsing behavior that can be categorized as: Saving page, Printing page, Adding page to Favorites, copying page content, the same page browsing times and page browsing time. Considering the five factors of user's browsing behaviors as well as the size of a page, an algorithm based on the speed of a page browsing was derived for the user interest degree. Meanwhile, a BP neural network was utilized for the fusion of the user interest degree. The training samples of the BP neural network were feed by collected user's browsing behaviors. The rationality and feasibility of the algorithm for the user interest degree was verified in the research. © 2005 - 2012 JATIT & LLS. All rights reserved.
Yu R.,Xian University of Finance and Economics
Advances in Intelligent and Soft Computing | Year: 2011
With the rapid development of multimedia and network technology, produces a new type of media - Streaming Media. Streaming Media thoroughly overcomes the defects that the traditional Internet can only show texts and images. It sets the video, audio and pictures at an organic whole and will become the mainstream of the Internet application in future. Based on the basic principle and characteristics of Streaming Media, this paper expounds the making of Streaming Media files and discusses applications and advantages of this technology in modern distance education. © 2011 Springer-Verlag Berlin Heidelberg.
Deng H.,Xian University of Finance and Economics
Advance Journal of Food Science and Technology | Year: 2015
Comprehend the importance of using different nutritional foods for different track and field items, so as to improve the competition ability and overall quality of track and field athletes in practical training in a better way. Mainly the method of literature reading and sorting, questionnaire survey and analysis summary are adopted. Investigate the using status of nutritional foods for different track and field items, analyze existing problems combining pertinent literature and summarize the influence of rational use of nutritional foods on different track and field items. Currently, use of nutritional foods for track and field athletes is rational. Physical quality of track and field athletes can only be improved by absorbing more nutritional dairy foods, as well as nutritional fruits and vegetables. © Maxwell Scientific Organization, 2015.
Luo Y.X.,Xian University of Finance and Economics
Journal of Communications | Year: 2015
Aiming at the problem of large overhead and low accuracy on the identification of obfuscated and malicious code, a new algorithm is proposed to detect malicious code by identifying multidimensional features based on ReliefF and Boosting techniques. After a disassembly analysis and static analysis for the clustered malicious code families, the algorithm extracts features from four dimensions: two static properties (operation code sequences and bytecode sequence) and two features (system call graph and function call graph) which combines the semantic features to reflect the behaviour characteristic of the malware, and then selects important feature vectors based on Relief. Finally, ensemble learning is carried out, and the decision result is boosted based on weighted voting according to accuracy for a different feature analysis. It has been proven by experiment and comparison that the algorithms have a much higher accuracy of the testing dataset with low overhead. © 2015 Journal of Communications.