检索结果-内蒙古大学图书馆

International Conference on Multimedia, Communication and computing Application, MCCA 2014

作者： Li, H.Y. Liu, R.F. Xu, W.R. Shi, R.S. School of Information and Communication Engineering Beijing University of Posts and Telecommunication Beijing China Education Ministry Key Laboratory of Trustworthy Distributed Computing and Service BUPT School of Humanities BUPT Beijing China

ISBN: (纸本)9781138027756

That the summary of an article can be automatically generated is an important text processing technique. The technique can automatically extract the most useful information of the text and present that information, which makes it easy to navigate and judge the importance to readers. This paper is based on a variety of texts and structure features. Supporting vector-machine classification method is used to identify summary sentences, which abandoned the traditional experience weight setting to optimize. The effective system that combines keywords density and document structure with the best weights is implemented for automatic summarization. The experimental results show that our Density and Structure Based Summary system improves the automatic summarization results. It’s better than the other two systems and has a big promotion on the precision, recall and F score. © 2015 Taylor & Francis Group, London.

关键词： Text processing

来源：评论

学校读者我要写书评

暂无评论

AABN: Anonymity Assessment Model Based on Bayesian Network With Application to Blockchain

引用

China Communications 2019年第6期16卷 55-68页

作者： Tianbo Lu Ru Yan Min Lei Zhimin Lin School of Software Engineering Beijing University of Posts and Telecommunications100876 BeijingChina Key Laboratory of Trustworthy Distributed Computing and Service(Beijing University of Posts and Telecommunications) Ministry of Education100876 BeijingChina Information Security Center School of Cyberspace SecurityBeijing University of Posts and TelecommunicationsBeijing 100876China

Blockchain is a technology that uses community validation to keep synchronized the content of ledgers replicated across multiple users,which is the underlying technology of digital currency like *** anonymity of blockchain has caused widespread *** this paper,we put forward AABN,an Anonymity Assessment model based on Bayesian ***,we investigate and analyze the anonymity assessment techniques,and focus on typical anonymity assessment *** the related concepts involved in the assessment model are introduced and the model construction process is described in ***,the anonymity in the MIX anonymous network is quantitatively evaluated using the methods of accurate reasoning and approximate reasoning respectively,and the anonymity assessment experiments under different output strategies of the MIX anonymous network are analyzed.

关键词： blockchain anonymity assessment bayesian network MIX

来源：评论

学校读者我要写书评

暂无评论

Two-Stream Convolutional Neural Network for Action Recognition in Radar 6th

Two-Stream Convolutional Neural Network for Action Recogniti...

引用

6th Conference on Signal and Information Processing, Networking and Computers, ICSINC 2019

作者： Zhou, Hao He, Yuan Huang, Hai Jing, Xiaojun Laboratory of Biological Recognition and Information Application Beijing University of Posts and Telecommunications Beijing China Key Laboratory of Trustworthy Distributed Computing and Service Ministry of Education Beijing University of Posts and Telecommunications Beijing100876 China

ISBN: (纸本)9789811541629

In the domain of human action classification, information obtained by optical equipment is used as the input source in most case. However, these optical devices are often affected by environmental factors, which bring some trouble to human motion recognition. The radar signal can avoid these situations, therefore the radar signal has been more and more significant part in this domain. In recent research, most researchers used the Micro-Doppler spectrogram of radar to classify the action. It only makes use of the time-frequency domain information, but ignores the spatial domain information. However, for the radar signal of the human motion, there should be some differences between the information in the time-frequency domain and the information in the other domain such as the space domain. In this case, we employed a CNN with two streams, which can use both Micro-Doppler spectrogram and the echo of the radar to recognize the action. In our architecture, there are two separate CNN processing streams for extracting the features of two stream of input, and the features are combined with a late network. The experiment proves that our method has better accuracy than the input of a single stream. © 2020, Springer Nature Singapore Pte Ltd.

关键词： Convolution

来源：评论

学校读者我要写书评

暂无评论

Hot event detection for social media based on keyword semantic information 4

Hot event detection for social media based on keyword semant...

引用

4th IEEE International Conference on Data Science in Cyberspace, DSC 2019

作者： Yu, Zexuan Wu, Xu Xie, Xiaqing Xu, Jin Key Laboratory of Trustworthy Distributed Computing and Service Ministry of Education China School of Cyberspace Security China Beijing University of Posts and Telecommunications Library Beijing University of Posts and Telecommunications Beijing China

ISBN: (纸本)9781728145280

Statistical features are commonly used to detect hot events in social media. However, these features cannot represent the semantic similarity of synonym expression. Aiming at this problem, a semantic keyword-based hot event detection method for social media is proposed in this paper, which includes a semantic keywords model to distinguish real-world events from hot topics. The clusters are found using Affinity Propagation with statistic features. Also, related discussions from different perspectives of the event are merged using word embedding. After that, news titles and posts of hot topics are employed to train the semantic keywords model. This model is used to distinguish real-world events from clusters. Experiments show the proposed method can find hot events from social media data and related posts for further research. © 2019 IEEE.

关键词： Semantics

来源：评论

学校读者我要写书评

暂无评论

A Multi-Objective Emergency Resource Scheduling Method Based on MOEA/D 19

A Multi-Objective Emergency Resource Scheduling Method Based...

引用

Proceedings of the 2019 2nd International Conference on Algorithms, computing and Artificial Intelligence

作者： Yanrong Li Xiaoyong Li Liyang Hou Wenping Kong Wenxue Ma Key Laboratory of Trustworthy Distributed Computing and Servic Beijing University of Posts and Telecommunications Beijing China National Engineering Laboratory For Integrated Command And Dispatch Technology Beijing China

ISBN: (纸本)9781450372619

After unconventional events occur, rapid emergency response is often needed in order to reduce losses. When there are more than one disaster points and more than one supply points, how to coordinate the local dispatch of emergency resources among these points is a problem worth studying. In this paper, we design a multi-objective, multi-point to multi-point and multi-resource emergency scheduling model that consider resources arrive time and transportation cost. Then a new emergency resource scheduling algorithm based on multi objective evolutionary algorithm based on decomposition (MOEA/D) is proposed, a new mutation operator was also designed to solve this problem. The experimental results show that this method can get a feasible scheduling scheme, which is faster than NSGA-II and has better scheduling results.

关键词： MOEA/D

来源：评论

学校读者我要写书评

暂无评论

A deepfake video detection method based on multi-modal deep learning method

A deepfake video detection method based on multi-modal deep ...

引用

作者： Yutong Zhang Xiaoyong Li Jie Yuan Yali Gao Linghui Li Key Laboratory of Trustworthy Distributed Computing and Service Beijing University of Posts and Telecommunications

Recently,most deepfake video classification tasks depend on frame-level features and try to train deep neural networks to characterize fake *** this kind of methods can achieve good results,they also waste the audio information and timing information of the deepfake video ***,to make better use of the audio information,we propose a multi-modal method to detect deepfake *** principle of our method is based on the mismatch between audio information and visual information,such as the inconsistency of mouth shape and *** calculate the modality dissonance score(MDS score) of videos to classify true/false *** experiments reach 84.4% accuracy on the DFDC dataset,which demonstrates the effectiveness of our method.

关键词： deepfake detection multi-modality adaptive modality dissonance deep learning deep neural network

来源：评论

学校读者我要写书评

暂无评论

Multi-objective Explanations of GNN Predictions

Multi-objective Explanations of GNN Predictions

引用

IEEE International Conference on Data Mining (ICDM)

作者： Yifei Liu Chao Chen Yazheng Liu Xi Zhang Sihong Xie Key Laboratory of Trustworthy Distributed Computing and Service (MoE) BUPT Lehigh University

ISBN: (纸本)9781665423991

Graph Neural Network (GNN) has achieved state-of-the-art performance in various high-stake prediction tasks, but multiple layers of aggregations on graphs with irregular structures make GNN a less interpretable model. Prior methods use simpler subgraphs to simulate the full model, or counterfactuals to identify the causes of a prediction. The two families of approaches aim at two distinct objectives, “simulatability” and “counterfactual relevance”, but it is not clear how the objectives can jointly influence the human understanding of an explanation. We design a user-study to investigate such joint effects, and use the findings to design a multi-objective optimization (MOO) algorithm to find Pareto optimal explanations that are well-balanced in simulatability and counterfactual. Since the target model can be of any GNN variants and may not be accessible due to privacy concerns, we design a search algorithm using zero-th order information without accessing the architecture and parameters of the target model. Quantitative experiments on nine graphs from four applications demonstrate that the Pareto efficient explanations dominate single-objective baselines that use first-order continuous optimization or discrete combinatorial search. The explanations are further evaluated in robustness and sensitivity to show their capability of revealing convincing causes, while being cautious about the possible confounders. The diverse dominating counterfactuals can certify the feasibility of algorithmic recourse, that can potentially promote algorithmic fairness where humans are participating in the decision-making using GNN.

关键词： Privacy Sensitivity Decision making Predictive models Pareto optimization Prediction algorithms Robustness

来源：评论

学校读者我要写书评

暂无评论

Deep learning based primary user classification in Cognitive Radios

Deep learning based primary user classification in Cognitive...

引用

International Symposium on Communications and Information Technologies (ISCIT)

作者： Yuanhao Cui Xiao jun Jing Songlin Sun Xiaohan Wang Dongmei Cheng Hai Huang Key Laboratory of Trustworthy Distributed Computing and service (BUPT) Beijing University of Posts and Telecommunications Beijing China Key Laboratory of Trustworthy Distributed Computing and service (BUPT) Beijing University of Posts and Telecommunications Beijing Beijing CN Key Laboratory of Trustworthy Distributed Computing and service (BUPT) School of Software and Microelectronics Beijing China

Deep Belief Networks (DBN) is a very powerful algorithm in deep learning. The DBN has been effectively applied in many areas of machine learning, such as computer vision (CV) and natural language processing (NLP). With the help of deep architecture, their accuracy has been largely improved and their human annotation data which traditional machine learning algorithm extremely rely on could be reduced. In Cognitive Radios (CRs), learning is necessary for its cognition, while two of the key challenges are how to classify primary user agents with their performances and predict their behaviors. The CRs' performance has a positive correlation with the hit rate of learning algorithm's classification and prediction results. In this paper, we study the questions of classification and prediction of user agents. We apply the DBN model to improve accuracy rating of user agent's recognition in CRs with user-centered model, it's the first application of deep learning structure in CRs. The DBN model provides a primary user agent's classification, which is the foundation of the prediction to both idle frequency spectrums and time slots. Experimental results show that the cognitive engine finds a much better detection rate than the CRs engine with shallow learning and other traditional strategy. The simulation results are also tested on the WIFI channel with 5GHz and 2.4GHz.

关键词： Training Machine learning algorithms Classification algorithms Prediction algorithms Machine learning Sensors Artificial neural networks

来源：评论

学校读者我要写书评

暂无评论

Research on improve topic representation over short text 3

Research on improve topic representation over short text

引用

3rd IEEE International Conference on Data Science in Cyberspace, DSC 2018

作者： Li, Lingyun Sun, Yawei Han, Xu Wang, Cong Key Laboratory of Trustworthy Distributed Computing and Service Beijing University of Posts and Telecommunications Ministry of Education Beijing100876 China School of Software Engineering Beijing University of Posts and Telecommunications Beijing100876 China

ISBN: (纸本)9781538642108

According to the characteristics of sparseness, poor focus and lack of semantic information in short text, the existing studies mainly improved the topic representation from two aspects: Adjusting the structure of the topic model to increase word co-occurrence and incorporating word embedding to enrich the semantic information. In this paper, we review the existing topic representation methods for short text from these two aspects, select DMM, BTM and LF-DMM as targets and compare the quality of topic representations from them. Considering the traditional topic model does not incorporate word embedding, and single short text often lacks context information due to length limitation, we try to combine it with word embedding, we use two real-world datasets to evaluate the quality of topic representation in the classification task: Although the LF-DMM model incorporates word embedding, it performs poorly on short text, and the performance of DMM and BTM integrated with word embedding improve greatly. © 2018 IEEE.

关键词： Embeddings

来源：评论

学校读者我要写书评

暂无评论

An AdamW-Based Deep Neural Network Using Feature Selection and Data Oversampling for Intrusion Detection

An AdamW-Based Deep Neural Network Using Feature Selection a...

引用

IEEE International Conference on Data Science in Cyberspace (DSC)

作者： Zhuoer Lu Xiaoyong Li Pengfei Qiu Key Laboratory of Trustworthy Distributed Computing and Service (BUPT) Ministry of Education Beijing China

With the development of the Internet and the increasing number of Internet users, cyber security has become a major concern for most netizens. In this paper, we propose an AdamW-based neural network using feature selection and data oversampling for intrusion detection. First, we use the Random Forest classifier to select the 25 most important features for classifying network traffic. Second, given the imbalance of different types of samples in the NSL-KDD dataset, we use ADASYN oversampling to oversample the minority samples. In addition, to achieve better performance, we use AdamW as the optimizer of our deep neural network. Finally, we tune the hyperparameters of our deep neural network to get the best classification results for intrusion detection. Compared with other classical machine learning models for intrusion detection, our deep neural network achieves high intrusion detection performance: the test set loss is reduced to 0.0001 and the test set accuracy is improved to 99.8%.

关键词：

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：