检索结果-内蒙古大学图书馆

2023 IEEE International Conference on Systems, Man, and Cybernetics, SMC 2023

作者： Yuan, Shuozhi Guo, Wenming Yang, Feng Beijing University of Posts and Telecommunications Key Laboratory of Trustworthy Distributed Computing and Service Beijing China Beijing University of Posts and Telecommunications Xinjiang Institute of Engineering Beijing China Xinjiang Institute of Engineering Urumqi China

ISBN: (纸本)9798350337020

Face detection is a fundamental and practical problem in computer vision, which aims to indicate the face positions in a wild environment precisely. However, different from the generic object detection tasks, there are a large number of face samples that suffer from unconstrained poses, occlusion, extreame lights, or other harmful conditions. YOLOV5 is an incredible milestone in the object detection area, but still not powerful enough for the challenging samples. To ease these difficulties, in this paper, we customize DSH-YOLOV5, a practical face detector. Specifically, we integrate a decoupled head with swin transformer layers, whose self-attention mechanism has great potential to explore subtle interconnection details. Additionally, we use two context modules (CBAM and SSH) to enhance the performance of features. Furthermore, we design a novel copy-paste data augmentation to fit the above challenging scenes. Extensive experiments demonstrate that we achieve SOTA performance on WIDER FACE, FDDB, and PASCAL FACE with competitive computational costs. Moreover, under COVID-19, we add the breathing mask and gender classification branches based on the ROI Align to produce more practical face information. © 2023 IEEE.

关键词： Face Detection Swin-Transformer YOLOV5

来源：评论

学校读者我要写书评

暂无评论

An Unsupervised Graph Embedding Method Based on Dynamic Graph Attention Networks and Infomax for Link Prediction

An Unsupervised Graph Embedding Method Based on Dynamic Grap...

引用

IEEE International Conference on Data Science in Cyberspace (DSC)

作者： Jingzhu Lu Feifan Song Jingchen Wu Xu Wu Key Lab of Trustworthy Distributed Computing and Service Ministry of Education School of Cyberspace Security Beijing University of Posts and Telecommunications Beijing China Key Lab of Trustworthy Distributed Computing and Service Ministry of Education School of Economics and Finance Changchun Finance College Jilin China Key Lab of Trustworthy Distributed Computing and Service Ministry of Education School of Computer Science Beijing University of Posts and Telecommunications Beijing China

Due to the lack of labeled data in many real-world scenarios, Graph Neural Network models that require a large volume of labeled data are not effective, and research based on unsupervised learning frameworks has become essential. To solve the problem of sparsity in labeled graph data, this paper proposes an unsupervised graph representation learning method based on graph infomax and dynamic graph attention mechanism(IMDGA). This method conducts unsupervised learning based on the idea of contrastive learning and optimizes the encoder by calculating the maximum mutual information between the graph's local information and global information. At the same time, we use a dynamic Graph Attention Network as an encoder model to prevent the Graph Convolutional Neural network's inability to adapt to neighbor nodes with different weights. Finally, link prediction experiments on three public data sets of cora, citeseer, and pumbed show that the proposed model has achieved superior results compared with other graph representation learning, showing its effectiveness and competitiveness.

关键词：

来源：评论

学校读者我要写书评

暂无评论

SSDAEE: Semi-Supervised Data-Augmented Event Extraction for Transfer Learning

SSDAEE: Semi-Supervised Data-Augmented Event Extraction for ...

引用

IEEE International Conference on Data Science in Cyberspace (DSC)

作者： Yueyue Shen Feifan Song Jingchen Wu Xu Wu Key Lab of Trustworthy Distributed Computing and Service Ministry of Education School of Cyberspace Security Beijing University of Posts and Telecommunications Beijing China Key Lab of Trustworthy Distributed Computing and Service Ministry of Education School of Economics and Finance Changchun Finance College Jilin China Key Lab of Trustworthy Distributed Computing and Service Ministry of Education School of Computer Science Beijing University of Posts and Telecommunications Beijing China

Although the general event extraction model has achieved good results in the standard data sets, there is a gap in the extraction results in low-resource environments. The general event extraction model still has room for improvement in the performance of special fields. Therefore, this paper proposes a semi-supervised reading comprehension event extraction model adapted to transfer learning, based on improved UDA for data enhancement, called SSDAEE, which can increase the accuracy of reading comprehension event type recognition. SSDAEE has a good performance on the public opinion data set. It can not only enhance a small number of supervised data in low-resource fields with unsupervised data but also improve the generalization ability of the model. Experiments show that when only 5% unsupervised data is used, the F1 score of SSDAEE on the public opinion data set is increased by 9% compared with the fully supervised model, and by 2% compared with other semi-supervised models.

关键词：

来源：评论

学校读者我要写书评

暂无评论

Blockchain Storage Method Based on Erasure Code 8

Blockchain Storage Method Based on Erasure Code

引用

8th International Conference on Data Science in Cyberspace, DSC 2023

作者： Meng, Fanyao Li, Jin Gao, Jiaqi Liu, Junjie Ru, Junpeng Lu, Yueming School of Cyberspace Security Beijing University of Posts and Telecommunications Key Lab of Trustworthy Distributed Computing and Service Ministry of Education Beijing China School of Information and Communication Engineering Beijing University of Posts and Telecommunications The State Key Laboratory of Information Photonics and Optical Communications Beijing China

ISBN: (纸本)9798350331035

Blockchain, as an emerging distributed database, effectively addresses the issue of centralized storage in IoT data, where storage capacity cannot match the explosive growth in devices and data scale, as well as the contradictions arising from centralized data management concerning data privacy and security concerns. To alleviate the problem of excessive pressure on single-point storage and ensure data security, a blockchain data storage method based on erasure codes is proposed. This method involves constructing mathematical functions that describe the data to split the original block data into multiple fragments and add redundant slices. These fragments are then encoded and stored in different locations using a circular hash space with the addition of virtual nodes to ensure load balancing among nodes and reduce situations where a single node stores too many encoded data blocks, effectively enhancing the storage space utilization efficiency of the distributed storage database. The blockchain storage method stores encoded data digest information such as storage location, creation time, and hashes, allowing for the tracing of the origin of encoded data blocks. In case of accidental loss or malicious tampering, this enables effective recovery and ensures the integrity and availability of data in the network. Experimental results indicate that compared to traditional blockchain approaches, this method effectively reduces the storage pressure on nodes and exhibits a certain degree of disaster recovery capability. © 2023 IEEE.

关键词： Blockchain Consistent Hashing Erasure Codes Internet of Things

来源：评论

学校读者我要写书评

暂无评论

An Intrusion Detection Method Based on Transformer-LSTM Model

An Intrusion Detection Method Based on Transformer-LSTM Mode...

引用

Neural Networks, Information and Communication Engineering (NNICE), International Conference on

作者： Zhipeng Zhang Xiaotian Si Linghui Li Yali Gao Xiaoyong Li Jie Yuan Guoqiang Xing Key Laboratory of Trustworthy Distributed Computing and Service Beijing University of Posts and Telecommunications Beijing China

With the development of network technologies, network intrusion has become increasing complex which makes the intrusion detection challenging. Traditional intrusion detection algorithms detect intrusion traffic through intrusion traffic characteristics or machine learning. These methods are inefficient due to the dependence of manual work. Therefore, in order to improve the efficiency and the accuracy, we propose an intrusion detection method based on deep learning. We integrate the Transformer and LSTM module with intrusion detection model to automatically detect network intrusion. The Transformer and LSTM can capture the temporal information of the traffic data which benefits to distinguish the abnormal data from normal data. We conduct experiments on the publicly available NSL-KDD dataset to evaluate the performance of our proposed model. The experimental results show that the proposed model outperforms other deep learning based models.

关键词： Measurement Deep learning Machine learning algorithms Intrusion detection Manuals Artificial neural networks Transformers

来源：评论

学校读者我要写书评

暂无评论

LMR-CBT: learning modality-fused representations with CB-Transformer for multimodal emotion recognition from unaligned multimodal sequences

引用

Frontiers of Computer Science 2024年第4期18卷 39-47页

作者： Ziwang FU Feng LIU Qing XU Xiangling FU Jiayin QI School of Computer Science(National Pilot Software Engineering School) Beijing University of Posts and TelecommunicationsBeijing 100876China Key Laboratory of Trustworthy Distributed Computing and Service(BUPT) Beijing 100876China Shanghai International School of Chief Technology Officer East China Normal UniversityShanghai 200062China School of Computer Science and Technology East China Normal UniversityShanghai 200062China School of Cyberspace Security Guangzhou UniversityGuangdong 510006China

Learning modality-fused representations and processing unaligned multimodal sequences are meaningful and challenging in multimodal emotion *** approaches use directional pairwise attention or a message hub to fuse language,visual,and audio ***,these fusion methods are often quadratic in complexity with respect to the modal sequence length,bring redundant information and are not *** this paper,we propose an efficient neural network to learn modality-fused representations with CB-Transformer(LMR-CBT)for multimodal emotion recognition from unaligned multi-modal ***,we first perform feature extraction for the three modalities respectively to obtain the local structure of the ***,we design an innovative asymmetric transformer with cross-modal blocks(CB-Transformer)that enables complementary learning of different modalities,mainly divided into local temporal learning,cross-modal feature fusion and global self-attention *** addition,we splice the fused features with the original features to classify the emotions of the ***,we conduct word-aligned and unaligned experiments on three challenging datasets,IEMOCAP,CMU-MOSI,and *** experimental results show the superiority and efficiency of our proposed method in both *** with the mainstream methods,our approach reaches the state-of-the-art with a minimum number of parameters.

关键词： modality-fused representations cross-model blocks multimodal emotion recognition unaligned multimodal sequences computational affection

来源：评论

学校读者我要写书评

暂无评论

An Abnormal Behavior Detection Method Based on User Behavior Correlation Feature Sequence Modeling 9

An Abnormal Behavior Detection Method Based on User Behavior...

引用

9th IEEE International Conference on Data Science in Cyberspace, DSC 2024

作者： Zhou, Yuting Sun, Lijuan Wu, Jingchen Gao, Yutong Wu, Xu Ministry of Education Key Laboratory of Trustworthy Distributed Computing and Service China Beijing University of Posts and Telecommunications School of Cyberspace Security BUPT Beijing China Beihang University Intellectual Property Information Service Center BHU Beijing China Minzu University of China School of Information Engineering MUC Beijing China

ISBN: (纸本)9798350391367

The healthcare industry is one of the industries most affected by insider threats. At present, the mainstream abnormal behavior detection of internal threats has the problems of relatively single reference dimension of user behavior data and lack of correlation analysis between data of different dimensions. Therefore, this paper proposes an abnormal behavior detection method that integrates user behavior correlation feature sequence modeling (UBCFSM). This method fully considers the multidimension and correlation of user behavior data, and uses the time convolutional network to learn behavior sequence features and the bidirectional long and short term memory network to learn behavior habit features of the convolutional attention mechanism. In addition, the deep neural network is introduced to fuse the deviations between the predicted results of different models and the actual results, detect abnormal results, and improve the generalization ability and detection accuracy of the model. Finally, experiments were carried out on the CERT public data set, and the AUC score of 97.1% was obtained. Compared with other abnormal behavior detection models, the proposed model has achieved significant improvement, showing its effectiveness and competitiveness. © 2024 IEEE.

关键词： Convolutional neural networks

来源：评论

学校读者我要写书评

暂无评论

An IoT Device Recognition Method based on Convolutional Neural Network

An IoT Device Recognition Method based on Convolutional Neur...

引用

Neural Networks, Information and Communication Engineering (NNICE), International Conference on

作者： Minghao Lu Linghui Li Yali Gao Xiaoyong Li Key Laboratory of Trustworthy Distributed Computing and Service Beijing University of Posts and Telecommunications Beijing China

Device recognition is the primary step toward a secure IoT system. However, the existing equipment recognition technology often faces the problems of unobvious data characteristics and insufficient training samples, resulting in low recognition rate. To address this problem, a convolutional neural network-based IoT device recognition method is proposed. We first extract the background icons of various IoT devices through the Internet, and then use the ResNet50 neural network to extract icon feature vectors to build an IoT icon library, and realize accurate identification of device types through image retrieval. The experimental results show that the accuracy rate of sampling retrieval in the icon library can reach 98.5%, and the recognition accuracy rate outside the library can reach 83.3%, which can effectively identify the type of IoT devices.

关键词： Training Face recognition Image retrieval Feature extraction Libraries Internet of Things Convolutional neural networks

来源：评论

学校读者我要写书评

暂无评论

Encrypted Traffic Classification Model Based on SwinT-CNN

Encrypted Traffic Classification Model Based on SwinT-CNN

引用

International Conference on Computer Engineering and Applications (ICCEA)

作者： Yueyang Wang Yali Gao Xiaoyong Li Jie Yuan Key Laboratory of Trustworthy Distributed Computing and Service Beijing University of Posts and Telecommunications Beijing China

With the rapid development of the network, the proportion of encrypted traffic in the network is increasing, which brings great challenges to traffic classification. The traditional encrypted traffic classification algorithm extracts the port number of the traffic or the load information of the traffic packet for feature matching, which has no effect in the current encrypted traffic. The encrypted traffic classification algorithm based on machine learning especially relies on the features extracted by professionals for encrypted traffic, which is inefficient and has low accuracy. And, The accuracy and overall performance of existing deep learning methods also need to be improved. Therefore, in order to improve the accuracy of encrypted traffic classification, this paper proposes an encrypted traffic classification method based on deep learning, which tries to combine the core modules of CNN and Swin Transformer with the encrypted traffic classification model to realize the identification of the application type of encrypted traffic. Among them, the CNN module is able to capture the local spatial information characteristics of encrypted traffic data. The multi-head attention mechanism of Swin Transformer can capture global information about the association between attributes of the data. The model is tested on the public ISCX VPN-nonVPN encryption dataset to evaluate the performance of the proposed model. Experimental results show that the accuracy of our proposed model reaches 96.7%. Compared with the other best-performing deep learning models, the proposed model improves the accuracy by 3.1% and the F1 value by 2.7%.

关键词：

来源：评论

学校读者我要写书评

暂无评论

Improving Knowledge Distillation for Federated Learning on Non-IID Data

Improving Knowledge Distillation for Federated Learning on N...

引用

IEEE International Conference on Big Data

作者： Zongyi Chen Sanchuan Guo Liyan Shen Xi Zhang Zhuonan Chang Key Laboratory of Trustworthy Distributed Computing and Service (MOE) Beijing University of Posts and Telecommunications China

Federated learning (FL) leverages knowledge from decentralized clients to train a global model in a privacy-preserving manner. One of the critical challenges in FL is the heterogeneity of client local data (i.e. non-IID data), which can lead to significant performance degradation. Federated Distillation (FD) can mitigate this issue by distilling the client predictions on the unlabeled public data into a global student model. However, a simple averaging of the client predictions without considering model confidences may lead to erroneous predictions. Moreover, it is challenging to rectify the predicted pseudo labels as no human labels are available in the public data. To address these issues, we propose a novel FD framework named FedUSL, aiming to distill a powerful global model on non-IID data. Specifically, we estimate the uncertainty of each client’s prediction on the unlabeled data and adopt a weighted ensemble of their predictions in consideration of uncertainties. We further rectify the global model predictions by a self-label reassigning method, without the requirement of manual labels. Extensive experiments on image and text tasks show that our proposal can achieve superior performance than state-of-the-art methods, and incurs no extra computing burden on the client side. The code is available at https://***/r/FedUSL-6338/.

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：