Time filter

Source Type

Zhu R.,Jiaxing University | Zhu R.,State Key Laboratory for Novel Software Technology | Guo B.,Jiaxing University
IASP 10 - 2010 International Conference on Image Analysis and Signal Processing | Year: 2010

Rapid development of various resources available on the Internet, the network just acts as a double-edged sword, which spreads useful and harmful information to the users simultaneously. In this paper, a novel Altering model for the erotic images embedded in web pages is proposed. It not only detects network eroticism, but also screens the web pages containing erotic images. For achieving a real-time processing in a dynamic network environment, the model is implemented by a BHO technique and its every part is designed on a cost-effectiveness analysis. The recognition accuracy of erotic images is improved significantly by Introducing color correction into an image Alter and adopting window mechanism into a text detector. Experimental results show that the proposed method obtains an excellent performance in terms of both efficiency and accuracy. ©2010 IEEE. Source

Chunlin L.,Wuhan University of Technology | Chunlin L.,State Key Laboratory for Novel Software Technology | Layuan L.,Wuhan University of Technology
Information Sciences | Year: 2014

This paper presents exploiting composition of mobile devices for maximizing user QoS under energy constraints in mobile grid. Mobile device service composition process includes two parts: mobile device service provisioning through device service market and mobile device resource allocation through device resource market. Interactions among mobile grid user agent, mobile device service agent and mobile device resource agent are mediated by means of market mechanisms. Mobile device service composition optimization maximizes the interests of mobile grid user agent, mobile device service agent and mobile device resource agent. Utility function is used to specify QoS requirement of mobile grid users and benefit of mobile device resource agents. The problem of services composition of mobile devices is formulated by utility optimization. The paper also presents a mobile grid services composition algorithm to maximize user QoS under energy constraint. In the simulation, the performance evaluation of proposed algorithm for services composition of mobile devices is conducted and compared with other related works. © 2014 Elsevier Inc. All rights reserved. Source

Chen L.,Shanghai Institute of Technology | Fan G.,State Key Laboratory for Novel Software Technology | Fan G.,East China University of Science and Technology | Liu Y.,Shanghai Institute of Technology
International Journal of Computational Science and Engineering | Year: 2016

In this paper, we address the fault tolerant service composition with particular attention to QoS. A fault tolerant strategy for improving the performance of service composition is proposed. The strategy is composed of invocation mechanism, synchronisation mechanism and exception mechanism. Petri nets are used to observe the behaviours of basic components, and to describe their interrelationship. The transaction attributes, reliability, and time of service are also articulated. The composition mechanism systematically integrates these schemas into a fault tolerant model. Based on this, the analysis technology and its enforcement method are proposed, which can guarantee the correct behaviour of service composition while meeting the required reliability. The related theories of petri nets help prove the correctness of the strategy. Finally, the method is realised by a simplified travel service and is tested by several experiments. © 2016 Inderscience Enterprises Ltd. Source

Xu J.,State Key Laboratory for Novel Software Technology | Yao Y.,State Key Laboratory for Novel Software Technology | Tong H.,Arizona State University | Tao X.,State Key Laboratory for Novel Software Technology | Lu J.,State Key Laboratory for Novel Software Technology
IJCAI International Joint Conference on Artificial Intelligence | Year: 2015

Recommender system has becomean indispensable component in many e-commerce sites. One major challenge that largely remains open is the cold-start problem, which can be viewed as an ice barrier that keeps the cold-start users/items from the warm ones. In this paper, we propose a novel rating comparison strategy (RAPARE) to break this ice barrier. The center-piece of our RAPARE is to provide a fine-grained calibration on the latent profiles of cold-start users/items by exploring the differences between cold-start and warm users/items. We instantiate our RAPARE strategy on the prevalent method in recommender system, i.e., the matrix factorization based collaborative filtering. Experimental evaluations on two real data sets validate the superiority of our approach over the existing methods in cold-start scenarios. Source

Yao Y.,State Key Laboratory for Novel Software Technology | Tong H.,Arizona State University | Xu F.,State Key Laboratory for Novel Software Technology | Lu J.,State Key Laboratory for Novel Software Technology
Proceedings of the ACM SIGKDD International Conference on Knowledge Discovery and Data Mining | Year: 2014

Community Question Answering (CQA) sites have become valuable platforms to create, share, and seek a massive volume of human knowledge. How can we spot an insightful question that would inspire massive further discussions in CQA sites? How can we detect a valuable answer that benefits many users? The long-term impact (e.g., the size of the population a post benefits) of a question/answer post is the key quantity to answer these questions. In this paper, we aim to predict the long-term impact of questions/answers shortly after they are posted in the CQA sites. In particular, we propose a family of algorithms for the prediction problem by modeling three key aspects, i.e., non-linearity, question/answer coupling, and dynamics. We analyze our algorithms in terms of optimality, correctness, and complexity. We conduct extensive experimental evaluations on two real CQA data sets to demonstrate the effectiveness and efficiency of our algorithms. © 2014 ACM. Source

Discover hidden collaborations