检索结果-内蒙古大学图书馆

International Conference on Computer Supported Cooperative Work in Design

作者： Jiawei Jiang Xiang Tian Yubing Han Chunyu Hu Bin Feng Jiguo Yu Key Laboratory of Computing Power Network and Information Security Ministry of Education Shandong Computer Science Center (National Supercomputer Center in Jinan) Qilu University of Technology (Shandong Academy of Sciences) Jinan China Shandong Provincial Key Laboratory of Industrial Network and Information System Security Shandong Fundamental Research Center for Computer Science Jinan China School of Information Science and Technology Taishan University Taian China School of Computer Science and Engineering University of Electronic Science and Technology of China Chengdu China Big Data Institute Qilu University of Technology Jinan China

ISBN: (数字)9798331513054

ISBN: (纸本)9798331513061

Owing to the easy deployment and mobile flexibility, Unmanned Aerial Vehicle (UAV) assisted Mobile Edge computing (MEC) has been deemed as one potential technology for handling the computation-intensive tasks at terminal devices (TDs). In this work, a MEC architecture assisted by UAVs is designed which achieves efficient offloading, computing, and downloading for tasks from multiple TDs via aerial to aerial collaboration of two UAVs. In this architecture, the task offloading process contains two parts, i.e., the offloading from TDs to a mobile UAV which flies around TDs, and the offloading from the mobile UAV to a hovering UAV which hovers in the air. The computing tasks from TDs will be divided into three parts allocated to the TDs themselves, and both two UAVs. Upon completion of computation, the computation results are downloaded to the TDs. The optimization objective is to seek for an optimal task division strategy to attain the weighted total energy consumption minimization for all devices. Since the formulated optimization problem is not convex, we develop a two-step iteration algorithm which jointly optimizes computing frequency, task allocation volume, as well as UAV's trajectory based on the method of block coordinate descent. Simulation results confirm the effectiveness and performance advantages of the designed algorithm.

关键词： Energy consumption Simulation Collaboration Computer architecture Autonomous aerial vehicles Minimization Resource management Jamming Trajectory optimization Radio spectrum management

来源：评论

学校读者我要写书评

暂无评论

A Patent Fast Recommendation Algorithm Based on User Requirements Combined with an Inverted Table

A Patent Fast Recommendation Algorithm Based on User Require...

引用

Advanced Information, Mechanical Engineering, Robotics and Automation (AIMERA), IEEE International Conference on

作者： Wan Zheng Qingyan Ding Na Li Yu Pan Xuecheng Dong Key Laboratory of Computing Power Network and Information Security Ministry of Education Shandong Computer Science Center (National Supercomputer Center in Jinan) Qilu University of Technology (Shandong Academy of Sciences) Jinan China Shandong Provincial Key Laboratory of Computer Networks Shandong Fundamental Research Center for Computer Science Jinan China

ISBN: (数字)9798350343335

ISBN: (纸本)9798350343342

Patent recommendation has always been an essential service for the innovation and development of all walks of life. The matching of demand and patent information can enable enterprises to quickly obtain the innovation and development trend of the industry. The current patent recommendation method has the problem of slow recommendation speed and insufficient comprehensive recommendation results. According to the existing problems, a patent fast recommendation algorithm based on user needs combined with inverted table is proposed. The algorithm using dual track recommended way for accurate recommendation and extensive recommendation, in the process of recommendation built by patent data inversion table and document list auxiliary recommendation to improve recommendation speed and accuracy, when accurate recommendation when using improved requirements based on GPT-2 compression model recommended after demand text compression, when widely recommended using TF-IDF mechanism for demand text of each paragraph after keyword extraction, finally generate widely recommendation list and accurate recommendation list. This recommendation method can not only realize the fast and accurate recommendation of patents, but also provide users with more innovative ideas for extensive recommendation. The experiments show that the proposed recommendation algorithm has higher speed and accuracy than the original recommendation algorithm.

关键词： Patents Vocabulary Technological innovation Accuracy Semantics Predictive models Writing Prediction algorithms Data mining Robotics and automation

来源：评论

学校读者我要写书评

暂无评论

OCF-HP: An Offline Compaction Framework Based on Hotspot Prediction for Data Lakehouse

OCF-HP: An Offline Compaction Framework Based on Hotspot Pre...

引用

Electronic technology and Information Science (ICETIS), International Conference on

作者： Bowen Sun Xiaohan Yang Ying Guo Zhihao Zhao Jing Chen Hu Zhang Jibin Wang Key Laboratory of Computing Power Network and Information Security Ministry of Education Shandong Computer Science Center (National Supercomputer Center in Jinan) Qilu University of Technology (Shandong Academy of Sciences) Jinan China Shandong Provincial Key Laboratory of Computer Networks Shandong Fundamental Research Center for Computer Science Jinan China

ISBN: (数字)9798350388343

ISBN: (纸本)9798350388350

With the rapid advancement of information technologies, the global data volume is increasing at an unprecedented speed. Data Lakehouse has become an effective solution for real-time aggregation of multimodal heterogeneous data and supporting multiple query modes. It is worth noting that due to the column-based storage commonly employed in Data Lakehouse and the resulting mechanisms such as Copy-on-Write and Merge-On-Read, data query speed, data freshness, and storage costs have become huge challenges. In this paper, an Offline Compaction Framework based on Hotspot Prediction (OCF-HP) was designed to address the above challenges. Specifically, we analyzed the Timeline data in Hudi which is the mainstream Data Lakehouse framework, then predicted the hotspot periods of data writing based on the improved LSTM model using such data, and realized automatic trigger of Hudi offline compaction tasks in reasonable non-hot periods. The experiment results show that OCF-HP provides Hudi with a more optimized offline compaction solution so that data lakehouse managers could achieve better query speed and data freshness with reducing storage costs in Hudi's read optimized mode.

关键词： Analytical models Costs Data integrity Predictive models Writing Data models Compaction

来源：评论

学校读者我要写书评

暂无评论

Android Malware Detection Method Based on Graph Convolutional networks

Android Malware Detection Method Based on Graph Convolutiona...

引用

Artificial Intelligence and Power Systems (AIPS), International Conference on

作者： Qingling Xu Shumian Yang Lijuan Xu Dawei Zhao Key Laboratory of Computing Power Network and Information Security Ministry of Education Shandong Computer Science Center (National Supercomputer Center in Jinan) (Shandong Academy of Sciences) Qilu University of Technology Jinan China Shandong Provincial Key Laboratory of Computer Networks Shandong Fundamental Research Center for Computer Science Jinan China

ISBN: (数字)9798350392210

ISBN: (纸本)9798350392227

Facing the serious security threats posed by Android malware amidst the rapid development of mobile internet, there is an urgent need to develop more accurate and efficient detection technologies. This paper introduces an innovative Android malware detection solution based on graph neural networks(GCN). The solution centers around function call graphs as core features, incorporating sensitive API identification and node importance analysis to streamline redundant information and reinforce key feature representation. Leveraging graph neural network models, complex call graphs are efficiently encoded into low-dimensional features, revealing the deep structure of malicious code. This method not only significantly improves detection accuracy to 99% but also drastically reduces detection time while ensuring performance, demonstrating its strong advantage in Android malware detection and classification tasks.

关键词： Training Accuracy Costs Graph convolutional networks Biological system modeling Feature extraction Malware Power systems Complexity theory security

来源：评论

学校读者我要写书评

暂无评论

SOD-YOLO: A New Small Object Traffic Sign Recognition network

SOD-YOLO: A New Small Object Traffic Sign Recognition Networ...

引用

International Conference on Computer Supported Cooperative Work in Design

作者： Xunjia Yin Xiaoming Wu Xiangzhi Liu Ministry of Education Shandong Computer Science Center (National Supercomputer Center in Jinan) Key Laboratory of Computing Power Network and Information Security Qilu University of Technology (Shandong Academy of Sciences) Jinan China Shandong Provincial Key Laboratory of Computer Networks Shandong Fundamental Research Center for Computer Science Jinan China

ISBN: (数字)9798350349184

ISBN: (纸本)9798350349191

Traffic sign recognition is a challenging task for unmanned systems, especially the problem of detecting small targets. During traffic sign recognition (TSR), traffic sign targets are small, represented by a small number of pixels in the image, and lack sufficient detail, making it difficult to detect them using a detector. To solve these problems a new traffic sign target detection algorithm is proposed, which introduces a new convolutional module SPD-Conv to capture the feature information of small target traffic signs more efficiently and replaces the C3 module in the backbone network with a C2f module, which greatly reduces the parameters and computation of the original network and achieves a lighter weight to make the model easier to deploy. A new weighted bi-directional feature pyramid network is used to replace the original feature pyramid network. In addition, a coordinate lightweight attention module is introduced for the small target feature loss problem, and finally, a new loss function is proposed to improve the convergence speed of the model while improving the small target detection accuracy. Extensive experimental results on CCTSDB and tt-100k datasets show that our method is more versatile and superior for traffic sign small target detection compared to several state-of-the-art methods.

关键词： YOLO Adaptation models Accuracy Target recognition Computational modeling Detectors Feature extraction

来源：评论

学校读者我要写书评

暂无评论

Multimodal Knowledge Graph Completion Method Based on Integrated Modality Adversarial Training and Relation-Enhanced Attention Mechanism

Multimodal Knowledge Graph Completion Method Based on Integr...

引用

International Conference on Computer Supported Cooperative Work in Design

作者： Ruijia Zhang Zheng Dong Jianqiang Zhang Gongpeng Song Qin Lu Key Laboratory of Computing Power Network and Information Security Ministry of Education Shandong Computer Science Center Qilu University of Technology (Shandong Academy of Sciences) Jinan China Shandong Engineering Research Center of Big Data Applied Technology Faculty of Computer Science and Technology Qilu University of Technology (Shandong Academy of Sciences) Jinan China Shandong Provincial Key Laboratory of Industrial Network and Information System Security Shandong Fundamental Research Center for Computer Science Jinan China Shandong Branch of China Mobile Communication Group Design Institute Co. Jinan China

ISBN: (数字)9798331513054

ISBN: (纸本)9798331513061

Negative sampling (NS) is widely used in knowledge graph completion (KGC) to generate negative triples for contrastive learning during training. However, existing NS methods are not suitable when multimodal information is incorporated into KGC models. Due to their complex design, these methods are also inefficient. In this paper, we propose the integration of Modality-Aware Adversarial Training (IMAT) to generate higher-quality negative samples for Multimodal Knowledge Graph Completion (MMKGC), and introduce a Relation-Enhanced Cross-modal Attention (RECA) mechanism to evaluate bidirectional attention weights between multimodal features using relational information, thereby improving the model's ability to identify hard negative samples. Our approach represents a joint design of MMKGC models and training strategies, surpassing 16 recent MMKGC methods and achieving new state-of-the-art results on three public MMKGC benchmarks.

关键词： Training Federated learning Design methodology Computational modeling Semantics Focusing Knowledge graphs Contrastive learning Sampling methods Real-time systems

来源：评论

学校读者我要写书评

暂无评论

Cross Auto-Encoder for Inscription Character Inpainting

Cross Auto-Encoder for Inscription Character Inpainting

引用

International Joint Conference on Neural networks (IJCNN)

作者： Long Zhao Zonglong Yuan Yuhao Lou Key Laboratory of Computing Power Network and Information Security Ministry of Education Shandong Computer Science Center (National Supercomputer Center in Jinan) Qilu University of Technology (Shandong Academy of Sciences) Jinan China Shandong Provincial Key Laboratory of Computer Networks Shandong Fundamental Research Center for Computer Science Jinan China

ISBN: (数字)9798350359312

ISBN: (纸本)9798350359329

Character inpainting, a challenging task, has seen slow development. Existing networks focus on either enhancing the encoding process, involving feature extraction or improving the decoding process, which revolves around character reconstruction. Some also work on refining the loss function for training the model. However, none of these models enhances both the encoding and decoding processes in character inpainting. In response to this gap, our paper leverages insights from the field of image inpainting, specifically the multi-stage inpainting network, to introduce the Cross Auto-Encoder. The Cross Auto-Encoder comprises parallel encoders and decoders that simultaneously enhance both the encoding and decoding processes in character inpainting. The two parallel encoders employ dilated convolution and branching convolution channel attention modules to extract character features concurrently. Meanwhile, the parallel decoder incorporates a self-attention mechanism, utilizing different loss functions for training, sharing parameters, and decoding the reconstructed character. Our extensive experiments on the expanded inscription dataset demonstrate that the Cross AutoEncoder surpasses other existing character inpainting models in the context of inscriptions. Furthermore, we conducted experiments on the handwritten Chinese character recognition dataset, yielding results superior to those achieved by current character inpainting models. In conclusion, the Cross Auto-Encoder stands out as a novel and effective approach, successfully addressing the shortcomings of existing character inpainting models by enhancing both encoding and decoding processes.

关键词： Training Convolution Refining Feature extraction Encoding Decoding Complexity theory

来源：评论

学校读者我要写书评

暂无评论

A Multi-Scale Infrared and Visible Image Fusion network Based on Context Perception

A Multi-Scale Infrared and Visible Image Fusion Network Base...

引用

International Conference on Computer Supported Cooperative Work in Design

作者： Huixuan Zhao Jinyong Cheng Rundong Du Key Laboratory of Computing Power Network and Information Security Ministry of Education Shandong Computer Science Center (National Supercomputer Center in Jinan) Qilu University of Technology (Shandong Academy of Sciences) Jinan China Shandong Provincial Key Laboratory of Computer Networks Shandong Fundamental Research Center for Computer Science Jinan China

ISBN: (数字)9798350349184

ISBN: (纸本)9798350349191

Existing image fusion methods often fail to fully utilize the contextual information of the images, resulting in blurry and distorted fusion results. In addition, although the existing multi-scale feature extraction methods can consider the information of the source image from multiple perspectives, they tend to ignore the local information. These limitations degrade the performance and effectiveness of image fusion techniques in practical applications. To address above issues, we propose a fusion network architecture called MIA-Nest, consisting of encoder network, fusion network, and decoder network. Specifically, we introduce the Squeeze-and-Excitation Blocks(SEBs) into the encoder network, so that it can take local information into account while extracting multi-scale features. At the same time, in order to allow the fusion network to integrate the context information of the extracted features, we introduce Global Context Blocks (GCBs), so as to enrich the details in the fusion result and enable it to contain more information. Finally, the decoder network reconstructs the fusion image from the fused features obtained by the fusion network. Experimental results demonstrate that our proposed method significantly improves the quality and detail preservation capability of the fused images.

关键词： Federated learning Computer architecture network architecture Feature extraction Decoding Data mining Image fusion

来源：评论

学校读者我要写书评

暂无评论

Revisiting the Application of Twin Connected Parallel networksand Regression Loss Functions in Industrial Defect Detection

SSRN

引用

SSRN 2023年

作者： Su, Zhanzhi Zhou, Mingle Li, Min Zhang, Zekai Zhang, Cheng Xing, Rui Li, Gang Key Laboratory of Computing Power Network and Information Security Ministry of Education Shandong Computer Science Center National Supercomputer Center in Jinan Qilu University of Technology Shandong Academy of Sciences Jinan250353 China

Industrial product surface defects can affect the reliability and lifetime of products. However, industrial scenes are complex, with a wide variety of products, and product surface defects often have characteristics such as large-scale variations and tiny targets, which makes it challenging to achieve the expected results in defect detection. To this end, this paper designs a Twin-Connected Parallel network (TCPNet) for high-precision defect detection and recognition. The innovative features of TCPNet include three points: First, a twin-connected parallel network is designed to efficiently extract semantic and texture information of objects through parallel networks, enhance important feature information of different branch networks in real-time, and ensure local-to-global feature extraction. Second, Differential Feature Remodeling Module(DFRM) modules designed. Carry out feature differentiation processing, and reshape the different features in the local feature extraction module to highlight the defective features. Third, Considering the difficulty of regression and model convergence for small objects, this paper designs the Dynamic Variation regression loss function (DVIoU) based on the dynamic distance relationship between the four pairs of vertices of the target frame and the real frame. Finally, this paper proposes an industrial defect dataset for bottlebody and practically applies TCPNet to industrial scenarios, with experimental results reaching SOTA. © 2023, The Authors. All rights reserved.

关键词： Surface defects

来源：评论

学校读者我要写书评

暂无评论

Lightweight Malicious Encrypted Traffic Detection Method Based on KL Image Compression

Lightweight Malicious Encrypted Traffic Detection Method Bas...

引用

IEEE International Conference on Data Science in Cyberspace (DSC)

作者： Yuecheng Wen Xiaotong Li Haiqing Lv Menglu Wang Wenbo Zuo Key Laboratory of Computing Power Network and Information Security Ministry of Education Shandong Computer Science Center (National Supercomputer Center in Jinan) Qilu University of Technology (Shandong Academy of Sciences) Jinan China Shandong Provincial Key Laboratory of Computer Networks Shandong Fundamental Research Center for Computer Science Jinan China

ISBN: (数字)9798350391367

ISBN: (纸本)9798350391374

Timely and accurate detection of and response to potentially malicious encrypted traffic is crucial in cybersecurity strategies. Convolutional neural network-based approaches that convert traffic loads into images for analysis have made significant progress in malicious encrypted traffic detection, but they suffer from the challenges of large number of participants, large network size, and high demand for computational resources, especially when edge computing devices are limited. Therefore, there is an urgent need for a lightweight, efficient and accurate malicious encrypted traffic detection model. This paper proposes a malicious encrypted traffic detection model whose core idea is to utilize KL image compression technique. The model first converts the raw traffic data into grayscale images and applies KL transform to compress the images efficiently. Subsequently, a convolutional neural network model with lightweight design is used to classify the compressed image for fast and accurate malicious encrypted traffic detection. After validating on both the public traffic dataset and the lab’s self-collected dataset, the results indicate that our method requires just 0.0216 seconds to process a single input sample, achieving an average classification accuracy of 98%. These findings highlight its practicality and excellence in addressing the real-world challenges of malicious encrypted traffic.

关键词： Image coding Accuracy Computational modeling Transforms Telecommunication traffic Gray-scale Data models Mobile handsets Cryptography Convolutional neural networks

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：