Entity

Time filter

Source Type


Lili C.,Taizhou Vocational and Technical College
International Journal on Smart Sensing and Intelligent Systems | Year: 2015

This article introduces ensemble learning algorithms in recommender systems, and in boosting algorithm framework of this article, shows how to filter the basic recommendation algorithm according to the characteristics of boosting algorithm. By comparing the rational choice of the two recommended boosting algorithm is applied to the frame. And then it determines the main parameters of the algorithm through the experiments, ultimately to obtain a more effective integration of the recommendation algorithm. Experimental results on Netflix validate the effectiveness of the proposed algorithm. Source


A highly specific and sensitive method for glucose quantification in human serum samples based on on-column enzymatic assay is described. In this method, the head of the capillary was used as a nanoliter-microreactor, the diluted samples spiked with a novel fluorogenic reagent named 2-[6-(4′-amino) phenoxy-3H-xanthen-3-on-9-yl] benzoic acid (APF), and the mixed enzyme solutions of glucose oxidase (GOx) and horseradish peroxidase (HRP), were individually injected into the capillary. Hydrogen peroxide (H2O2) generated in situ by catalytic reaction between GOx and glucose, activates APF in the presence of HRP to form a highly fluorescent product, which was electrophoretically separated from the unreacted APF and detected by the LIF detector. The proposed method allowed the determination of glucose down to 10 nM in real samples, with RSD values lower than 3.5%, which also has the potential for measurements of multicomponents in many other systems including measurement of α-glucosidase activity and screening for its inhibitors. © 2016 WILEY-VCH Verlag GmbH & Co. KGaA, Weinheim. Source


Xu Q.,Taizhou Vocational and Technical College
2010 International Conference on Intelligent Computation Technology and Automation, ICICTA 2010 | Year: 2010

User profiling technology is discussed based on the diversity and randomness of user demands. The key technologies, which include profile initialization and updating, are researched. The user profile is expressed in form of vector space model. It's built using centroid-based classification method. User's browsing behaviors are analyzed to get feedback implicitly, calculating the documents degrees. A periodic adaptive learning mechanism based on Rocchio algorithm is put forward. The profile's capability to dynamic track user needs is verified using satisfaction as evaluation indicator in experiments. © 2010 IEEE. Source


He J.H.,Taizhou Vocational and Technical College
Applied Mechanics and Materials | Year: 2013

This paper focuses on the hierarchical control system of previously developed robotic stingray that consists of PC, DSP and bottom controller, achieved direction control and velocity control of the biomimetic stingray like propulsor. Finally, an experiment was carried out to investigate the influence of fin kinematic parameters (such as frequency, amplitude and wavelength) on propulsion speed of the stingray, followed by variance analysis. The experimental results illustrate that the biomimetic stingray like propulsor works well under the control of the hierarchical control system. © (2013) Trans Tech Publications, Switzerland. Source


Li K.J.,Taizhou Vocational and Technical College
Advanced Materials Research | Year: 2013

The effects of Gd on microstructure of AZ71 magnesium alloy were studied by optical microscope, SEM and X-ray diffraction analyzer. The results show that the alloy mainly contained α-Mg matrix, eutectic α(Mg)+β(Mg17Al12) phase and Al2Gd phase. The microstructure of alloy was refined and homogeneous by Gd. The volume fraction of β phase is decreased with the addition of Gd, and the β-phase are congregated and coarsened with the excessive Gd. © (2013) Trans Tech Publications, Switzerland. Source

Discover hidden collaborations