检索结果-内蒙古大学图书馆

IEEE International Conference on Multimedia and Expo (ICME)

作者： Xiaoqian Liu Xiuyun Li Yuan Cao Fan Zhang Xiongnan Jin Jinpeng Chen School of Computer Science (National Pilot Software Engineering School) Beijing University of Posts and Telecommunications Beijing China Key Laboratory of Trustworthy Distributed Computing and Service (BUPT) Ministry of Education Beijing China The Technology Innovation Center of Cultural Tourism Big Data of Hebei Province Chengde China Hebei Normal University for Nationalities Chengde China Knowledge Discovery and Data Mining Research Center Zhejiang Lab Hangzhou China

Next-POI recommendation aims to explore from user check-in sequence to predict the next possible location to be visited. Existing methods are often difficult to model the implicit association of multi-modal data with user choices. Moreover, traditional methods struggle to fully explore the variation of user preferences at variable time intervals. To tackle these limitations, we propose a Multi-Modal Temporal Knowledge Graph-aware Sub-graph Embedding approach (Mandari). We first construct a novel Multi-Modal Temporal Knowledge Graph. Based on the proposed knowledge graph, we integrate multi-modal information and leverage the graph attention network to calculate sub-graph prediction probability. Next, we implement a temporal knowledge mining method to model the segmentation and periodicity of user check-in and obtain temporal prediction probability. Finally, we fuse temporal prediction probability with the previous sub-graph prediction probability to obtain the final result. Extensive experiments demonstrate that our approach outperforms existing state-of-the-art methods.

关键词：

来源：评论

学校读者我要写书评

暂无评论

Chinese open relation extraction with pointer-generator networks 5

Chinese open relation extraction with pointer-generator netw...

引用

5th IEEE International Conference on Data Science in Cyberspace, DSC 2020

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

ISBN: (纸本)9781728195582

Most of the traditional Chinese open relation extraction (Open RE) system exploit the syntactic, lexical and other language structure information obtained by natural language processing (NLP) tools from sentences to build hand-crafted patterns for extraction, which is easy to cause error propagation and affect the accuracy of extraction. In this paper, we propose an end-to-end abstract Chinese Open RE model based on the Pointer-Generator network, PGCORE. We employ the results extracted by the state-of-the-art pattern-based Chinese Open RE system as the training set of the model. Experimental results show that our method is outperforms the pattern-based several baselines system, which proves the feasibility and effectiveness of using deep learning models for Chinese Open RE. © 2020 IEEE.

关键词： Extraction

来源：评论

学校读者我要写书评

暂无评论

RDI: An adversarial robustness evaluation metric for deep neural networks based on sample clustering features

arXiv

引用

arXiv 2025年

作者： Song, Jialei Zuo, Xingquan Wang, Feiyang Huang, Hai Zhang, Tianle Shool of Computer Science Beijing University of Posts and Telecommunications Beijing China Key Laboratory of Trustworthy Distributed Computing and Service Ministry of Education Beijing China Shool of Cyberspace Security Beijing University of Posts and Telecommunications Beijing China

Deep neural networks (DNNs) are highly susceptible to adversarial samples, raising concerns about their reliability in safety-critical tasks. Currently, methods of evaluating adversarial robustness are primarily categorized into attack-based and certified robustness evaluation approaches. The former not only relies on specific attack algorithms but also is highly time-consuming, while the latter due to its analytical nature, is typically difficult to implement for large and complex models. A few studies evaluate model robustness based on the model’s decision boundary, but they suffer from low evaluation accuracy. To address the aforementioned issues, we propose a novel adversarial robustness evaluation metric, Robustness Difference Index (RDI), which is based on sample clustering features. RDI draws inspiration from clustering evaluation by analyzing the intra-class and inter-class distances of feature vectors separated by the decision boundary to quantify model robustness. It is attack-independent and has high computational efficiency. Experiments show that, RDI demonstrates a stronger correlation with the gold-standard adversarial robustness metric of attack success rate (ASR). The average computation time of RDI is only 1/30 of the evaluation method based on the PGD attack. Our open-source code is available at: https://***/r/RDI-B1DA. Copyright © 2025, The Authors. All rights reserved.

关键词： Open systems

来源：评论

学校读者我要写书评

暂无评论

Robust Text Clustering with Graph and Textual Adversarial Learning

Robust Text Clustering with Graph and Textual Adversarial Le...

引用

IEEE International Conference on Data Science in Cyberspace (DSC)

作者： Yongming Liang Tian Tian Kaifu Jin Xiaoyu Yang Yuefei Lv Xi Zhang Key Laboratory of Trustworthy Distributed Computing and Service Ministry of Education Beijing University of Posts and Telecommunications Beijing China

ISBN: (数字)9781728195582

ISBN: (纸本)9781728195599

Text clustering is a fundamental task that finds groups of similar texts in the corpus. Deep learning based models can capture the semantic and syntactic information in local word sequences to represent the texts but may ignore global text structures in the corpus. Recently, Graph Convolutional Networks (GCNs) have achieved the state-of-the-art performance in many applications, which may facilitate the text clustering task by exploiting the graph structures between texts via words and phrases. However, GCNs are vulnerable to adversarial attacks, e.g., small perturbations in graph structures and node attributes lead to poor performance. To address this issue, we propose a Robust Text Clustering (RTC) approach which consists of both autoencoder and GCN modules to exploit the local text features and global text structures respectively in one unified framework. We generate two types of adversarial perturbations targeting text structure and graph structure, and develop an adversarial training method to improve the robustness and generalization of the RTC framework. Extensive experiments on two real-world datasets demonstrate the effectiveness of our proposal in text clustering.

关键词：

来源：评论

学校读者我要写书评

暂无评论

Real-time event detection and tracking in microblog via text chain and sentiment time series 5

Real-time event detection and tracking in microblog via text...

引用

5th IEEE International Conference on Data Science in Cyberspace, DSC 2020

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

ISBN: (纸本)9781728195582

Microblog becomes one of the most popular media for sharing first-hand information. As more close to real event, detecting and tracking event in microblogs is a research hotspot. To alleviate feature sparsity problem, traditional event detection and tracking methods exploit comments, replies, and reposts to enrich the information of posts. However, ignoring text relationship only leads to more irrelevant information. Aiming at this issue, a novel real-time event detection and tracking method for microblog stream is proposed. For each microblog from the real-time stream, similarity against historical event set is first evaluated. Then, known event microblogs are partitioned by the proposed method sentiment time series, clustered based on word relation graph, and compare with the previous graph to track burst keywords. Meanwhile, text chain is introduced to detect events for unknown event microblogs. Text chain is a unique relationship of microblogs, in which comments, replies, and reposts are used to form a word relation graph of the corresponding post. Posts and attached graphs are clustered to detect new events. Experiments are performed on real-time microblog datasets to evaluate the effectiveness of methods. Results show that the proposed method gains better F-measure than similar methods. © 2020 IEEE.

关键词： Time series

来源：评论

学校读者我要写书评

暂无评论

An efficient intrusion detection model based on deepFM

An efficient intrusion detection model based on deepFM

引用

IEEE Information Technology, Networking, Electronic and Automation Control Conference

作者： Yuchen Ji Xiaoyong Li Ministry of Education Key Laboratory of Trustworthy Distributed Computing and Service (BUPT) Beijing University of Posts and Telecommunication Beijing China

ISBN: (数字)9781728143903

ISBN: (纸本)9781728143910

With the development of the Internet, more and more network attacks are also emerging. The traditional intrusion detection system has been challenged tremendously, and it is unable to detect the new types of attacks. This paper proposes an intrusion detection model based on deepFM, which combines the advantages of FM algorithm for processing shallow features and the advantages of deep neural network for processing deep features, which can effectively extract features and classify them. We then performed experiments using the KDD CUP 99 dataset. First, the training set is deduplicated and oversampled to balance various types of samples, and then the processed data is used for training. Then optimize tuning parameters and find an optimal model. Finally, use the best model to make predictions on the test set. On the test set, we got a correct rate of 0.934 and a COST value of 0.191. By analyzing the results and comparing existing studies, we can conclude that the model proposed in this paper has a good effect on KDD 99 CUP and has certain practical value.

关键词： Feature extraction Intrusion detection Data models Encoding Frequency modulation Neural networks Training

来源：评论

学校读者我要写书评

暂无评论

EML-Based IM/DD Transmission over Dispersion-Uncompensated Links Using Low-Complexity Optimized Detection

EML-Based IM/DD Transmission over Dispersion-Uncompensated L...

引用

Asia Communications and Photonics Conference and Exhibition (ACP)

作者： Shuangyue Liu Han Cui Xizi Tang Yueming Lu Yaojun Qiao State Key Laboratory of Information Photonics and Optical Communications School of Information and Communication Engineering Beijing University of Posts and Telecommunications Beijing China Key Laboratory of Trustworthy Distributed Computing and Service Ministry of Education School of Cyberspace Security Beijing University of Posts and Telecommunications Beijing China

We experimentally demonstrate an EML-based 40-Gbit/s OOK IM/DD trans-mission over 120-km SSMF using a low-complexity optimized detection algorithm, which requires no multiplication during the update of survival path, ... 详细信息

ISBN: (纸本)9781957171005

关键词： Chromatic dispersion Asia Optical fiber communication Complexity theory Detection algorithms Photonics

来源：评论

学校读者我要写书评

暂无评论

A High Accuracy DNS Tunnel Detection Method Without Feature Engineering

A High Accuracy DNS Tunnel Detection Method Without Feature ...

引用

International Conference on Computational Intelligence and Security

作者： Yang Chen XiaoYong Li Beijing Key Laboratory of Trustworthy Distributed Computing and Service (Beijing University of Posts and Telecommunications) Ministry of Education Beijing China

ISBN: (纸本)9781665404457;9781665447119

Domain Name System (DNS) is a key protocol and service used on the Internet. It is responsible for converting domain names into IP addresses. DNS tunnel is a method of encoding data of other programs or protocols in DNS query and response. Previous studies usually need to extract a large number of features manually and train the classifier of DNS tunnel detection by feature engineering. In this paper, a new framework for DNS tunnel detection is proposed, which can automatically extract features, including long short-term memory (LSTM) language model with attention mechanism and gated recurrent unit (GRU) language model with attention mechanism. Finally, a single-level classifier based on a character-level convolutional neural network (Char-CNN) is proposed. The results show that the LSTM and GRU language models based on attention mechanism and the algorithm of character-level convolution neural network achieve high accuracy and near-zero false positives.

关键词： Protocols Neural networks Training data Detectors Logic gates Feature extraction Classification algorithms

来源：评论

学校读者我要写书评

暂无评论

Multi-level Directed Fuzzing for Detecting Use-after-Free Vulnerabilities

Multi-level Directed Fuzzing for Detecting Use-after-Free Vu...

引用

IEEE International Conference on Trust, Security and Privacy in computing and Communications (TrustCom)

作者： Yuntao Zhang Zhongru Wang Weiqiang Yu Binxing Fang Key Laboratory of Trustworthy Distributed Computing and Service Ministry of Education Beijing University of Posts and Telecommunications Beijing China Chinese Academy of Cyberspace Studies Beijing China Beijing DigApis Technology Co. Ltd Beijing China

ISBN: (纸本)9781665416597

Greybox fuzzing has been widely used in vulnerabilities detection. Most greybox fuzzing tools are coverage-based, which usually use basic block transition to gain code coverage and focus on improving it to trigger more bugs. However, only increasing code coverage is insufficient to find some heap-based vulnerabilities such as use-after-free (UAF) and double-free (DF). This is because, to trigger these vulnerabilities, one needs not only to cover more code, but also to execute special heap operations to satisfy a particular temporal constraint (i.e., allocating heap memory, free memory, and accessing the heap memory). In this paper, we propose an approach, namely MDFuzz, to detect heap-based vulnerabilities adopting multi-level directed greybox fuzzing. The key idea is identifying different targets to guide the fuzzing process to cover specific heap operations without wasting resources exploring unrelated program components. We first perform a static analysis to automatically recognize three critical targets related to heap operations and then calculate each basic block's distance to the targets. Moreover, we propose a probability-based multi-level seed queue and a novel seed selection strategy to augment the guidance of directed fuzzing. To evaluate MDFuzz, we have performed an evaluation on 7 real-world applications. The experimental results demonstrate that MDFuzz significantly outperforms the state-of-the-art fuzzers, including AFL, AFLFast and VUzzer, in terms of the time consumed to discover heap-based vulnerabilities. Moreover, MD-Fuzz found 4 previously unknown vulnerabilities in real-world programs.

关键词： Privacy Target recognition Conferences Computer bugs Static analysis Binary codes Fuzzing

来源：评论

学校读者我要写书评

暂无评论

Scalable Blind Matching: An Efficient Ciphertext Matching Scheme for Content-Based Pub/Sub Cloud services

Scalable Blind Matching: An Efficient Ciphertext Matching Sc...

引用

IEEE International Conference on Big Data

作者： Qingling Feng Ruisheng Shi Qifeng Luo Lina Lan Jinqiao Shi Key Laboratory of Trustworthy Distributed Computing and Service (BUPT) Ministry of Education Beijing University of Posts and Telecommunications Beijing China

ISBN: (数字)9781728162515

ISBN: (纸本)9781728162522

Content-based publish/subscribe cloud services are prevailing recently. Confidentiality in publish/subscribe cloud services has become a major concern, especially for applications with sensitive data. Many methods have been proposed to achieve the confidentiality of events and subscriptions. Unfortunately, all these approaches suffer significant performance deterioration while matching on large-scale subscription set. In this paper, we propose an efficient ciphertext matching scheme called SBM. To the best of our knowledge, SBM is the first approach which can preserve confidentiality while at the same time provide scalable event matching. Furthermore, we have integrated SBM with an open source publish/subscribe middleware, PADRES and conducted extensive experiments to evaluate our scheme. The experimental results demonstrate that the matching speed of our solution is faster by two orders of magnitude than its counterparts.

关键词： Degradation Protocols Scalability Big Data Security Indexes Middleware

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：