Search

In this paper, we present a novel weighted version of semi-supervised discriminant analysis method by assigning weights to each labeled samples. The proposed within-class weight can detect the outliers and between-class weight can discover the support points in boundaries between different classes. In addition, our proposed method is robust to diverse-density classes and imbalanced boundaries. For high-dimensional dataset, our method can find a nice low-dimensional projection to preserve the discriminative information and manifold structure embedded in both labeled and unlabeled samples. It can also be easily kernelized to form a nonlinear method and do semi-supervised induction. The experiments show that our method can achieve very promising classification accuracies than other methods.

Cardiovascular disease has seriously affected the health of the elderly, and treatment of sudden cardiovascular disease is particularly difficult in most private pension institutions. A kind of portable ECG monitor is introduced in this paper which can be real-time, effective detection and treatment the patient's ECG information. It is beneficial for mastering the patient's ECG timely and building distinctive and intelligent private pension institution, and constantly improve its service quality and attractiveness to meet increasingly prominent and diverse pension service requirements. It is an important means to positive respond to the aging and alleviate the contradiction between supply and demand of pension services. The core of new portable ECG monitor is MCU based on ARM Cortex-M3 and its storage medium is a high capacity SD Card. Besides the device realizes user-machine interaction, waveform replay, arrhythmia analysis and illness alarm. This system can detect QRS wave and analyze in real-time and embed file system, so the ECG data can be stored in text form and readability and translatability of data are enhanced. For the MIT-BIH database and actual persons, the accuracy of monitor reaches the requirements.

According the diameter measurement of single crystal silicon in the growth process to put forward a diameter detection method that a circle defined by three points not in the same straight line. First, pretreat the growth image of single crystal silicon. Then, according the algorithm that a circle defined by three points not in the same straight line to extract the feature points, and the diameter pixel value of single crystal silicon accuratly detected. Finally transformed diameter pixel value into length value of the monocrystalline silicon rod through the measurement of diameter pixel value of draft tube. Tests show that error is very small between the test results and the true value, and meet the system requirements in a certain precision.

With rapid growth of the Internet and the globalization of the market, companies accepted and adopted the new information and communication technology in the performance of their activities, not only to support traditional activities, but also to support those arising from new opportunities, mainly from the Internet. This paper addresses the e-service quality issue in the electronic marketplace. The purpose of the paper is to investigate e-service quality dimensions from both e-service providerpsilas and customerpsilas perspectives. The paper explores e-service quality dimensions based on a review of the development of e-service quality dimension. It proposes a 10-dimension scale for measuring e-service quality: Website design, reliability, fulfillment, security, responsiveness, personalization, information and empathy from the e-service providerpsilas perspective, and trust and experience from the customerpsilas perspective. This paper concludes by discussing the limitation of this study and highlighting areas for the future research in the realm of e-service quality.

With a rapid deployment of mobile devices, mobile learning gives rise to new possibilities for extending learning opportunities to all social-economic levels. Nevertheless, current research on mobile learning has mostly been aimed at enhancing learning of school or college students. In this light, the paper seeks to throw light on the potential of mobile learning for distance learner communities, including problem teenagers, social employees and ageing people. Rather than being employed as a complementation to the current conventional learning and teaching scenarios, mobile learning tends to make more sense when it serves as an effective conduit for a particular learner community to access training and education. Also, mobile learning is of increasing importance when used to support the learning activities of hard-to-reach groups to underpin social transformation and to deal with the challenges posed by demographic shifts.