Institute of Electronic System Engineering

Beijing, China

Institute of Electronic System Engineering

Beijing, China
Time filter
Source Type

Chen J.,Beihang University | Wu Z.,Beihang University | Gao H.,Beihang University | Zhang C.,Beihang University | And 3 more authors.
Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics) | Year: 2013

In this paper, we aim to explore interesting landmark recommendations based on geo-tagged photos for each user. Meanwhile, we also try to answer such a question, i.e., when we want to go sightseeing in a large city such as Beijing, where should we go? To achieve our goal, first, we present a data field clustering method (DFCM). By using DFCM, we can cluster a large-scale geo-tagged web photo collection into groups (or landmarks) by location. And then, we provide more friendly and comprehensive overviews for each landmark. Subsequently, we model the users' dynamical behaviors using the fusion user similarity, which not only captures the overview semantic similarity, but also extract the trajectory similarity and the landmark trajectory similarity. Finally, we propose a personalized landmark recommendation algorithm based on the fusion user similarity. Experimental results show that our proposed approach can obtain a better performance than several state-of-the-art methods. © 2013 Springer-Verlag.

Chen J.,Beihang University | Liu Y.,Beihang University | Li D.,Institute of Electronic System Engineering
International Journal of Uncertainty, Fuzziness and Knowlege-Based Systems | Year: 2015

The recommender systems community is paying great attention to diversity as key qualities beyond accuracy in real recommendation scenarios. Multifarious diversity-increasing approaches have been developed to enhance recommendation diversity in the related literature while making personalized recommendations to users. In this work, we present Gaussian Cloud Recommendation Algorithm (GCRA), a novel method designed to balance accuracy and diversity personalized top-N recommendation lists in order to capture the user's complete spectrum of tastes. Our proposed algorithm does not require semantic information. Meanwhile we propose a unified framework to extend the traditional CF algorithms via utilizing GCRA for improving the recommendation system performance. Our work builds upon prior research on recommender systems. Though being detrimental to average accuracy, we show that our method can capture the user's complete spectrum of interests. Systematic experiments on three real-world data sets have demonstrated the effectiveness of our proposed approach in learning both accuracy and diversity. © 2015 World Scientific Publishing Company.

Chen J.,Beihang University | Gao H.,Beihang University | Wu Z.,Beihang University | Li D.,Institute of Electronic System Engineering
Proceedings of the International Conference on Parallel and Distributed Systems - ICPADS | Year: 2013

In this work, we address a novel problem about tag co-occurrence relationship prediction across heterogeneous networks. Although tag co-occurrence has recently become a hot research topic, many studies mainly focus on how to produce the personalized recommendation leveraging the tag co-occurrence relationship and most of them are considered in a homogeneous network. So far, few studies pay attention to how to predict tag co-occurrence relationship across heterogeneous networks. In order to solve the aforementioned problem, we propose a novel two-step prediction approach. First, weight path-based topological features are systematically extracted from the network. Then, a supervised model is used to learn the best weights associated with different topological features in deciding the co-occurrence relationships. Experiments are performed on real-world dataset, the Flickr network, with comprehensive measurements. Experimental results demonstrate that weight path-based heterogeneous topological features have substantial advantages over commonly used link prediction approaches in predicting co-occurrence relations in information networks. © 2013 IEEE.

Han Y.,Beihang University | Li D.,Beihang University | Li D.,Institute of Electronic System Engineering | Wang T.,Beihang University
Frontiers of Computer Science in China | Year: 2011

There has been considerable interest in designing algorithms for detecting community structure in real-world complex networks. A majority of these algorithms assume that communities are disjoint, placing each vertex in only one cluster. However, in nature, it is a matter of common experience that communities often overlap and members often play multiple roles in a network topology. To further investigate these properties of overlapping communities and heterogeneity within the network topology, a new method is proposed to divide networks into separate communities by spreading outward from each local important element and extracting its neighbors within the same group in each spreading operation. When compared with the state of the art, our new algorithm can not only classify different types of nodes at a more fine-grained scale successfully but also detect community structure more effectively. We also evaluate our algorithm using the standard data sets. Our results show that it performed well not only in the efficiency of algorithm, but also with a higher accuracy of partition results. © 2010 Higher Education Press and Springer-Verlag Berlin Heidelberg.

Liu Y.-C.,Tsinghua University | Liu Y.-C.,Institute of Electronic System Engineering | Ma Y.-T.,Institute of Electronic System Engineering | Ma Y.-T.,Wuhan University | And 3 more authors.
International Journal of Automation and Computing | Year: 2011

With the development of Internet technology and human computing, the computing environment has changed dramatically over the last three decades. Cloud computing emerges as a paradigm of Internet computing in which dynamical, scalable and often virtualized resources are provided as services. With virtualization technology, cloud computing offers diverse services (such as virtual computing, virtual storage, virtual bandwidth, etc.) for the public by means of multi-tenancy mode. Although users are enjoying the capabilities of super-computing and mass storage supplied by cloud computing, cloud security still remains as a hot spot problem, which is in essence the trust management between data owners and storage service providers. In this paper, we propose a data coloring method based on cloud watermarking to recognize and ensure mutual reputations. The experimental results show that the robustness of reverse cloud generator can guarantee users' embedded social reputation identifications. Hence, our work provides a reference solution to the critical problem of cloud security. © 2011 Institute of Automation, Chinese Academy of Sciences and Springer-Verlag Berlin Heidelberg.

Ma Y.-T.,Wuhan University | Ma Y.-T.,Institute of Electronic System Engineering | He K.-Q.,Wuhan University | Li B.,Wuhan University | And 2 more authors.
Journal of Computer Science and Technology | Year: 2010

Large-scale object-oriented (OO) software systems have recently been found to share global network characteristics such as small world and scale free, which go beyond the scope of traditional software measurement and assessment methodologies. To measure the complexity at various levels of granularity, namely graph, class (and object) and source code, we propose a hierarchical set of metrics in terms of coupling and cohesion - the most important characteristics of software, and analyze a sample of 12 open-source OO software systems to empirically validate the set. Experimental results of the correlations between cross-level metrics indicate that the graph measures of our set complement traditional software metrics well from the viewpoint of network thinking, and provide more effective information about fault-prone classes in practice. © 2010 Springer Science+Business Media, LLC & Science Press, China.

Huang Y.,Beijing Institute of Technology | Zhang Y.,Institute of Electronic System Engineering | Xia H.,Institute of Electronic System Engineering | Liang T.,Institute of Electronic System Engineering | Cheng D.,Beijing Institute of Technology
Proceedings of the IEEE Conference on Decision and Control | Year: 2012

Despite significant progress in the optimal theory of impulsive control systems, finding the optimal solution for them still remains a challenging task because of the computational complexity. In this paper, we focus our interests on the LQ-based optimization for a specific class of linear impulsive control systems mixed with continuous-time controls and fixed-time impulses so that the problem-solving ideas can be borrowed from the intensively studied and highly mature linear quadratic optimization theory, and the difficulty encountered in the conventional hybrid optimal control theory is bypassed because the impulsive instants are prescribed a priori. Using the classical Bellman Dynamic Programming, a matrix Riccati hybrid equation for the LQ-based optimization problem is derived and its steady-state solution is analyzed. The hybrid-type Riccati equation is formed by concatenating the matrix Riccati differential equation and the difference counterpart. Furthermore, the time-invariant system only with uniformly timing impulses is considered. In this case, the matrix Riccati hybrid equation is degenerated into a difference one, which is related to the discretized continuous-time dynamics. Finally, a simple regulator problem with impulsive control is given to validate the feasibility of the designed optimal feedback impulse control law. © 2012 IEEE.

Zhang H.,PLA University of Science and Technology | Gan W.,PLA University of Science and Technology | Xu F.,Institute of Electronic System Engineering
Proceedings - 9th Web Information Systems and Applications Conference, WISA 2012 | Year: 2012

We study the integration of individuals attributes to infer their influence ability in social network in this paper. The influence between individuals is usually asymmetric and can propagate via edges gradually. We suggest an Influence Factor Graph(IFG) which can integrate different node and edge features into a uniform inferring model. And for each node the model can compute personalized influence ability value. Experiment results in Zarchary and Wikipedia co-editing social networks show that, the model can depict influence reasonably and reveal some interesting social influence rules. © 2012 IEEE.

He Z.-Q.,PLA University of Science and Technology | Wu L.-F.,PLA University of Science and Technology | Zhang H.-S.,Institute of Electronic System Engineering | Li H.-B.,PLA University of Science and Technology | Lai H.-G.,PLA University of Science and Technology
Sichuan Daxue Xuebao (Gongcheng Kexue Ban)/Journal of Sichuan University (Engineering Science Edition) | Year: 2011

A novel approach based on semantics was proposed to express and match the security supply-and-demand policy of web service. Through constructing a general security ontology, the definition method and matching algorithm of semantic security supply-and-demand policy for web service were presented, and the matching problem of policy was translated into the subsumption reasoning problem of semantic concept. Both the theoretical analysis and experimental evaluation showed that the proposed approach can present the necessary semantic information in the representation of policy and effectively improve the accuracy of matching result, thus overcomes the deficiency of the syntactic approaches. This approach can also simplify the definition and administration of the policy at the same time, which thereby provides a more effective solution for the expression and matching problem of security policy in web service environment.

Li D.,Institute of Electronic System Engineering
Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics) | Year: 2011

The success of a search engine in cloud computing environment relies on the numbers of users and their click-through. If we take the previous search key words as tags of users to study and differentiate the user interaction behaviors, the search engine is able to actively push related and useful information to users based on their previous actions instead of passively waiting for users' queries. However the user searching behavior is affected by lots of factors, and it is quite complex and uncertain. The log files provided by a search engine have recorded all the information of the user interaction process on their servers or browsers, such as key words, click-through rate, time stamp, time on page, IP address, browser type and system stats, even the user location etc, which are all important information to understand and categorize users' searching behavior. Is there any statistical property almost independent to search key words? How to push recommendation based on the queried key words? And how to extract user behavior models of searching actions in order to recommend the information to meet users' real needs more timely and precisely? © 2011 Springer-Verlag Berlin Heidelberg.

Loading Institute of Electronic System Engineering collaborators
Loading Institute of Electronic System Engineering collaborators