Evolving Recommender System for Mobile Apps

Xiao Xia,Xiaodong Wang,Xingming Zhou 한국산학기술학회 2013 SmartCR Vol.3 No.3

The explosive growth of mobile apps gives rise to the significant challenge of app discovery. For this reason, Google Play utilizes a collaborative filtering approach for recommending apps to users by analyzing user behavior. Those recommendations help users discover apps by referring to the experiences of other users. However, their choices may also be limited because most users only know about a limited number of apps. To eliminate such constraints, we propose a novel recommendation method utilizing global information about apps. We generate recommendations by both analyzing the metadata and measuring the similarity between apps, leveraging the Latent Semantic Index method. To understand both methods, we further measure the diversity within them. Through those measurements, we not only gain better understanding of both recommendation methods but also discover new knowledge about user preferences. Such measurements also identify the necessity and potential to evolve the existing system. We therefore propose a diversity measurement?based evolution framework for the development of mobile app recommender systems. To implement the framework, we further model the system evolution as a multi-criteria optimization problem and design a rank aggregation scheme to solve it. Preliminary evaluations verified the promising effectiveness of our framework and method based on a data set of 103,348 apps.

The Automatic Recognition of Large Ball Valve Sealing Bolt Based on Digital Image

Qingjun Song,Xingming Xiao,Haiyan Jiang 보안공학연구지원센터 2015 International Journal of Signal Processing, Image Vol.8 No.6

In this paper, we put forward an effective algorithm for edge detection based on combination of multi-shape/scale Mathematical morphology and wavelet transformation. Firstly, the bimodal method and multi-threshold method were combined for the image segmentation. Then, the image edge was detected by the algorithm. Finally, the angle between the bolt and manipulator was calculated using the rotation conversion matrix. With the systematic error and correlation coefficient, the calculated angle was verified. Experimental results show that the method can protect the edge integrity of the nut image, with fast processing speed and strong anti-noise ability. The work in this paper provides a theoretical basis for the automatic recognition in the large ball valve sealing bolt system.

A new multi-class classification method based on minimum enclosing balls

QingJun Song,XingMing Xiao,HaiYan Jiang,XieGuang Zhao 대한기계학회 2015 JOURNAL OF MECHANICAL SCIENCE AND TECHNOLOGY Vol.29 No.8

With respect to classification problems, the Minimum enclosing ball (MEB) method was recently studied by some scholars as a newsupport vector machine. As a nascent technology, however, MEB reports poor adaptability for different types of samples, especiallymulti-class samples. In this paper, we propose a new multi-class classification method based on MEB. This method is derived from eachclass sample center and radius with the Gaussian kernel width factor parameter σ, which is labelled as σ-MEB. σ is a variable parameteraccording to the different sample characteristics. When this parameter is considered, the multi-class classifier is easy to adapt and is robustin diverse datasets. The quadratic programming problem was transformed into its dual form with Lagrange multipliers using thismethod. Finally, we applied sequential minimal optimization method and Karush—Kuhn—Tucker conditions to accelerate the trainingprocess. Numerical experiment results indicate that given different types of samples, the proposed method is more accurate than themethods with which it is compared. Moreover, the proposed method reports values in the upper quantile with respect to adaptation performance.

An Efficient Frame-Level Rate Control Algorithm for High Efficiency Video Coding

( Yubei Lin ),( Xingming Zhang ),( Jianen Xiao ),( Shengkai Su ) 한국인터넷정보학회 2016 KSII Transactions on Internet and Information Syst Vol.10 No.4

In video coding, the goal of rate control (RC) is not only to avoid the undesirable fluctuation in bit allocation, but also to provide a good visual perception. In this paper, a novel frame-level rate control algorithm for High Efficiency Video Coding (HEVC) is proposed. Firstly a model that reveals the relationship between bit per pixel (bpp), the bitrate of the intra frame and the bitrate of the subsequent inter frames in a group of pictures (GOP) is established, based on which the target bitrate of the first intra frame is well estimated. Then a novel frame-level bit allocation algorithm is developed, which provides a robust bit balancing scheme between the intra frame and the inter frames in a GOP to achieve the visual quality smoothness throughout the whole sequence. Our experimental results show that when compared to the RC scheme for HEVC encoder HM-16.0, the proposed algorithm can produce reconstructed frames with more consistent objective video quality. In addition, the objective visual quality of the reconstructed frames can be improved with less bitrate.

Study on the Preparation Process and Properties of Magnetorheological Fluid Treated by Compounding Surfactants

Xiangfan Wu,Xingming Xiao,Zuzhi Tian,Fei Chen 한국자기학회 2016 Journal of Magnetics Vol.21 No.2

Aiming to prepare high performance magnetorheological fluid, firstly, oleic acid and sodium dodecyl benzene sulfonate are chosen as surfactants. And then, the mechanical stirring process including stirring time, stirring temperature and stirring speed are optimized by measuring sedimentation ratio and zero-field viscosity. Finally, the properties of prepared magnetorheological fluid are elaborated. The results indicate that the compounding of oleic acid and sodium dodecyl benzene sulfonate can improve the properties of magnetorheological fluid distinctively, and the optimistic compounding content is 4g:4g or 5g:5g. The surfactants adding orders and the second stirring time have little effect on the properties of magnetorheological fluid, while obviously of the first stirring time, temperature and speed. Moreover, the sedimentation ratio of prepared magnetorheological fluid is less than 5.2% in two weeks, the zero-field viscosity is smaller than 0.6 Pa · s at 20℃, and the maximum yield stress is higher than 50 kPa.

Effect of Particle Characteristics and Temperature on Shear Yield Stress of Magnetorheological Fluid

Xiangfan Wu,Xingming Xiao,Zuzhi Tian,Fei Chen,Wang Jian 한국자기학회 2016 Journal of Magnetics Vol.21 No.2

Aiming to improve the shear yield stress of magnetorheological fluid, magnetorheological fluids with different particle characteristics are prepared, and the influence rules of particle mass fraction, particle size, nanoparticles content and application temperature on shear yield stress are investigated. Experimental results indicate that shear yield stress increases approximate linearly with the enhancement of particle mass fraction. Particle size and the nanoparticles within 10% mass fraction can improve the shear yield stress effectively. When the application temperature is higher than 100℃, the shear yield stress decreases rapidly because of thermal expansion and thermal magnetization effect.

Influence of Temperature on Torque Transmission Stability of Magnetorheological Fluid

Zuzhi Tian,Xiangfan Wu,Xingming Xiao,Li Hao Korean Magnetics Society 2018 Journal of Magnetics Vol.23 No.4

Aiming to obtain the influence mechanism of temperature on the torque transmission stability of magnetorheological fluid, firstly, a magnetorheological transmission device and its test-bed are established, and then, the thermal expansion, temperature pressure effect, heat volatilization, viscosity temperature characteristic, magnetic temperature characteristic and other related temperature characteristics were experimented systematically. The results indicate that the thermal expansion rate of 30 wt% magnetorheological fluid is 6 % and 18 % respectively at 120°C and 220°C, and the working space pressure is 13 kPa when the temperature rises from 32°C to 63°C. The thermal volatilization rate is still lower than 1 % at 200°C for 8 hours. The viscosity of the silicone oil based magnetorheological fluid decreases by 78 % when the temperature increases from 20°C to 240°C, and the saturation magnetization decreased by 6.2 %, 13 % and 23 % respectively at 150°C, 250°C and 350°C. Furthermore, the temperature field distribution experiment shows that the highest temperature region is at the outer diameter of the disk and expands slowly along the radial direction. Moreover, compared with rotational speed, temperature is the main factor affecting the torque fluctuation of magnetorheological fluid.