检索结果-内蒙古大学图书馆

A new framework for anomaly detection based on knn-distort in the metro traffic flow 6th

6th International Conference on Swarm Intelligence, ICSI 2015 held in conjunction with the 2nd BRICS Congress on Computational Intelligence, CCI 2015

作者： Zhang, Yapeng Hu, Nan Wang, Wenjun Yu, Hao School of Computer Science and Technology Tianjin University Tianjin300072 China Tianjin Emergency Medical Center Tianjin300072 China Tianjin Key Laboratory of Cognitive Computing and Application School of Computer Science and Technology Tianjin University Tianjin300072 China

ISBN: (纸本)9783319204710

Anomaly detection is an important problem that has been well researched in diverse application domains. However, to the best of our knowledge, the anomaly detection for metro traffic flow has not been investigated before. In this paper, we proposed a new framework to solve two problems about anomaly detection in the metro traffic flow: obtaining the potential information by every passenger’s trip and detecting anomalies among metro traffic flow. For the first problem, we proposed a novel encoding path model to infer the passing stations’ information for each trip. For the second problem, we provide an improved K-Nearest Neighbor Distort (KNN-Distort) algorithm to quantify the anomalies in the metro traffic flow. We conduct intensive experiments on a large real-world metro dataset to demonstrate the performance of our algorithms. © Springer International Publishing Switzerland 2015.

关键词： Encoding (symbols)

来源：评论

学校读者我要写书评

暂无评论

S-Looper: Automatic summarization for multipath string loops 2015

S-Looper: Automatic summarization for multipath string loops

引用

24th International Symposium on Software Testing and Analysis, ISSTA 2015

作者： Xie, Xiaofei Liu, Yang Le, Wei Li, Xiaohong Chen, Hongxu Tianjin Key Laboratory of Cognitive Computing and Application Tianjin University School of Computer Science and Technology China Nanyang Technological University Singapore Singapore Iowa State University United States

ISBN: (纸本)9781450336208

Loops are important yet most challenging program constructs to analyze for various program analysis tasks. Existing loop analysis techniques mainly handle well loops that contain only integer variables with a single path in the loop body. The key challenge in summarizing a multiple-path loop is that a loop traversal can yield a large number of possibilities due to the different execution orders of these paths located in the loop;when a loop contains a conditional branch related to string content, we potentially need to track every character in the string for loop summarization, which is expensive. In this paper, we propose an approach, named S-Looper, to automatically summarize a type of loops related to a string traversal. This type of loops can contain multiple paths, and the branch conditions in the loop can be related to string content. Our approach is to identify patterns of the string based on the branch conditions along each path in the loop. Based on such patterns, we then generate a loop summary that describes the path conditions of a loop traversal as well as the symbolic values of each variable at the exit of a loop. Combined with vulnerability conditions, we are thus able to generate test inputs that traverse a loop in a specific way and lead to exploitation. Our experiments show that handling such string loops can largely improve the buffer overflow detection capabilities of the existing symbolic analysis tool. We also compared our techniques with KLEE and PEX, and show that we can generate test inputs more effectively and efficiently. © 2015 ACM.

关键词： Symbolic execution String constraints Loop summarization

来源：评论

学校读者我要写书评

暂无评论

Exploring link-based algorithm for web spam detection

引用

Journal of Information and Computational Science 2015年第13期12卷 5003-5011页

作者： Yu, Jian Zhou, Jing Yu, Mei Du, Yu Lv, Fang Tianjin Key Laboratory of Cognitive Computing and Application School of Computer Science and Technology Tianjin University Tianjin China

Searching engines are playing more and more important role in discovering information on the web nowadays. Spam web, however, is one of the problems which reduce the efficiency of search engines. A more reasonable consideration is the reversibility of information transition in random walk model. In this paper, we propose TDRank algorithm which is on the basis of the two-direction transmission of information. Several good pages and spam pages are selected to be seeds in this paper. Their trust scores are propagating through both incoming links and outgoing links. Experiment results show that TDRank is effective for combating spam. Besides, a method of seed selection is put forward, seeds selected by our method help the results be more accurate. ©, 2015, Journal of Information and Computational Science. All right reserved.

关键词： Search engines

来源：评论

学校读者我要写书评

暂无评论

Dynamic resource provision for cloud broker with multiple reserved instance terms 1

引用

15th International Conference on Algorithms and Architectures for Parallel Processing, ICA3PP 2015

作者： Zhang, Jiangtao Chen, Shi Huang, Hejiao Wang, Xuan Du, Dingzhu School of Computer Science and Technology Harbin Institute of Technology Shenzhen Graduate School Shenzhen518055 China Public Service Platform of Mobile Internet Application Security Industry Shenzhen518057 China Shenzhen Key Laboratory of Internet of Information Collaboration Shenzhen518055 China Shenzhen Applied Technology Engineering Laboratory for Internet Multimedia Application Shenzhen518055 China Department of Computer Science University of Texas at Dallas RichardsonTX75080 United States

ISBN: (数字)9783319271194

ISBN: (纸本)9783319271187

Relying on the knowledge of the pricing benefit of long-term reserved resource and multiplexing gains, cloud broker strives to minimize its cost by utilizing infrastructure resources from public cloud service provider. Different reserved instance terms accompanied by different prices are provisioned by the provider. How to choose the appropriate ones from various terms to meet the dynamic user demands at the least cost is a great challenge. This paper addresses the challenge by two algorithms. Extensive real world traces driven evaluations show that the heuristic algorithm runs about twice as fast as the approximation one, while both algorithms can save almost the same resource cost up to 27%. © Springer International Publishing Switzerland 2015.

关键词： Costs

来源：评论

学校读者我要写书评

暂无评论

An efficient incremental algorithm for dynamic network community discovery

引用

Journal of Computational Information Systems 2015年第12期11卷 4313-4320页

作者： Yu, Mei Zhou, Jing Yu, Jian Du, Yu Ning, Jie Zhang, Xu Tianjin Key Laboratory of Cognitive Computing and Application School of Computer Science and Technology Tianjin University Tianjin China

Many community discovery methods need to divide dynamic social network of each moment. They often lead to high time complexity. There is a need for detecting community techniques on dynamic social networks with a high degree of accuracy and a low time complexity. In this study, we propose a FNICD (Fast-Newman based Incremental Community Detection) method, and compare it with the Incremental Clustering algorithm on test datasets. The experiment results reveal that our method has lower time complexity and detects more accurate community structures. In addition, the method defines the deviation degree to reduce cumulative errors. ©, 2015, Binary Information Press. All right reserved.

关键词： Complex networks

来源：评论

学校读者我要写书评

暂无评论

Brain network modeling analysis of the cross-modal integration modulated by auditory attention

Brain network modeling analysis of the cross-modal integrati...

引用

第十三届全国人机语音通讯学术会议(NCMMSC2015)

作者： Junhai Xu Yidan Sun Hongyan Wang Gaoyan Zhang Baolin Liu Jianwu Dang School of Computer Science and Technology Tianjin University Tianjin Key Laboratory of Cognitive Computing and Application

Researchers have put more attention on the functional brain networks. Under the influence of the human genome, scientists in brain science have developed the human brain connectome, which greatly promoted the study of human brain function network. Although, there are many brain network studies in auditory attention or visual attention, the mechanisms of cross-modal integration remain unclear. Our study is based on inter-modal semantic integration. To explore the neuronal patterns in specific cognitive conditions, we designed the EEG experiment based on our previous study. We applied the brain network modeling to the EEG data in order to explore the cross-modal integration modulated by spatial attention. In our study, we collected EEG data from 16 healthy subjects, using a combined auditory spatial attention and cross-modal priming paradigm. Subjects are constructed to see the picture on the screen, and concentrate on the one side(left ear or right ear), and then hear two voices, divided into left and right, then they judged whether the sound and picture in the attention side are consistent or not, and pressed the button to make the appropriate judgments. Then we constructed the functional brain networks in different conditions: Attention/no-attention, semantically-congruent/semantically-incongruent. The topological properties were calculated for each subject, and then the statistical analysis was conducted to explore the difference between distinct conditions. Our study found that the connectivity in no-attention condition was stronger, and the small-worldness for semantically-congruent condition was larger, which suggested that the capacity of information processing was more effective. Moreover, the correlation analysis discovered that there was a significant correlation between the local efficiency of functional brain networks and their behavioral performance. In conclusion, the brain network modeling analysis in the cross-modal integration modulated by auditory a

关键词： Auditory attention Cross-modal semantics Brain network modeling EEG

来源：评论

学校读者我要写书评

暂无评论

Brain network modeling analysis of the audio-visual information integration of the Chinese polyphonic character

Brain network modeling analysis of the audio-visual informat...

引用

第十三届全国人机语音通讯学术会议(NCMMSC2015)

作者： Shasha Fu Gaoyan Zhang Junhai Xu Jianwu Dang School of Computer Science and Technology Tianjin University Tianjin Key Laboratory of Cognitive Computing and Application

In the study of Chinese language, many previous studies investigated the cognitive mechanism of homophone, homograph and synonymy. However, the audio-visual information integration mechanism of the polyphone is still unclear. In this study, the electroencephalogram(EEG) signals were collected from 16 subjects when they read and listened to the polyphone simultaneously. Based on the theory of complex network, we built the EEG functional network, then calculated the network indexs and their correlation with the behaviors. Results showed that the human brain was a small-world network, and there was a negative correlation of the betweenness of T3 and Cz nodes with the reaction time. Based on the previous studies, we infer that T3 which occupys the temporal lobe may participate in the process of semantic information extraction, and Cz that occupys the central area may be associated with the integration and cognition of language.

关键词： Polyphone brain network modeling audio-visual information integration EEG

来源：评论

学校读者我要写书评

暂无评论

Content-based social network user interest tag extraction

International Journal of Database Theory and Application

引用

International Journal of Database Theory and application 2015年第2期8卷 107-118页

作者： Yu, Mei Han, Xu Gou, Xiaolu Yu, Jian Lv, Fang Li, Jingyu Tianjin Key Laboratory of Cognitive Computing and Application School of Computer Science and Technology Tianjin University Tianjin China

As a typical social network media, microblog has attracted a lot of users and quantities of information. More and more researchers and scholars are keen to explore useful information contained in microblog data and discover features of user interest for personalized recommendation. According to the unique feature of microblog text, this paper adjusts the traditional processes of Chinese word segmentation and tag extraction. Aiming to establish microblog user interest model, this paper combines clustering and classification algorithm to extract user interest tags. Experimenting upon Sina Weibo text dataset, the conclusion is reached that the method proposed in this paper is more effective and accurate to discover user features and the tags extracted are more in line with the user interest. © 2015 SERSC.

关键词： Data mining

来源：评论

学校读者我要写书评

暂无评论

Exploration of Image Search Results Quality Assessment

引用

IEEE Transactions on Big Data 2015年第3期1卷 95-108页

作者： Tian, Xinmei Lu, Yijuan Stender, Nate Yang, Linjun Tao, Dacheng CAS Key Laboratory of Technology Geo-Spatial Information Processing and Application Systems University of Science and Technology of China Hefei230027 China Department of Computer Science Texas State University San MarcosTX78666 United States Department of Computer Science University of Nebraska-Lincoln LincolnNE68588 United States Microsoft Corporation RedmondWA98052-7329 United States Centre for Quantum Computation and Intelligent Systems Faculty of Engineering and Information Technology University of Technology Sydney UltimoNSW2007 Australia

Image retrieval plays an increasingly important role in our daily lives. There are many factors which affect the quality of image search results, including chosen search algorithms, ranking functions, and indexing features. Applying different settings for these factors generates search result lists with varying levels of quality. However, no setting can always perform optimally for all queries. Therefore, given a set of search result lists generated by different settings, it is crucial to automatically determine which result list is the best in order to present it to users. This paper aims to solve this problem and makes four main innovations. First, a preference learning model is proposed to quantitatively study and formulate the best image search result list identification problem. Second, a set of valuable preference learning related features is proposed by exploring the visual characters of returned images. Third, a query-dependent preference learning model is further designed for building a more precise and query-specific model. Fourth, the proposed approach has been tested on a variety of applications including reranking ability assessment, optimal search engine selection, and synonymous query suggestion. Extensive experimental results on three image search datasets demonstrate the effectiveness and promising potential of the proposed method. © 2015 IEEE.

关键词： Image retrieval

来源：评论

学校读者我要写书评

暂无评论

Discriminative multi-view feature selection and fusion

Discriminative multi-view feature selection and fusion

引用

IEEE International Conference on Multimedia and Expo (ICME)

作者： Yanbin Liu Binbing Liao Yahong Han School of Computer Science and Technology Tianjin Universtiy China College of Computer Science and Technology Zhejiang University China Tianjin Key Laboratory of Cognitive Computing and Application China

In computer vision tasks such as action recognition and image classification, combining multiple visual feature sets is proven to be an effective strategy. However, simply combing these features may cause high dimensionality and lead to noises. Feature selection and fusion are common choices for multiple feature representation. In this paper, we propose a multi-view feature selection and fusion method which chooses and fuses discriminative features from multiple feature sets. For discriminative feature selection, we learn the selection matrix W by the minimization of the trace ratio objective function with ℓ 2,1 norm regularization. For multiple feature fusion, we incorporate local structures of each view in the Laplacian matrix. Since the Laplacian matrix is constructed in unsupervised manner and scaled category indicator matrix is solved iteratively, our work is fully unsupervised. Experimental results on four action recognition datasets and two large-scale image classification datasets demonstrate the effectiveness of multi-view feature selection and fusion.

关键词： Laplace equations Accuracy Support vector machines Linear programming Optimization Feature extraction Training

来源：评论

学校读者我要写书评

暂无评论

建议与咨询留下您的常用邮箱和电话号码，以便我们向您反馈解决方案和替代方法

时间限定

文献类型

馆藏选择

核心期刊

语言

文献类型

帮助

文字说明：

检索规则说明：

检索范例：

分类表

所选分类

限定检索结果

文献类型

馆藏范围

日期分布

学科分类号

主题

机构

作者

语言

请选择保存的检索档案：

请选择收藏分类：

通借通还

建议与咨询 留下您的常用邮箱和电话号码，以便我们向您反馈解决方案和替代方法

时间限定

文献类型

馆藏选择

核心期刊

语言

文献类型

帮助

文字说明：

检索规则说明：

检索范例：

分类表

所选分类

限定检索结果

文献类型

馆藏范围

日期分布

学科分类号

主题

机构

作者

语言

请选择保存的检索档案： 新增检索档案 确定 取消

请选择收藏分类： 新增自定义分类 确定 取消

通借通还

建议与咨询留下您的常用邮箱和电话号码，以便我们向您反馈解决方案和替代方法

请选择保存的检索档案：

请选择收藏分类：