检索结果-内蒙古大学图书馆

International Conference on Artificial Intelligence and Smart Environment, ICAISE 2022

作者： El Arby Chrif, Mohamed El Moustapha Saleck, Moustapha Mohamed N’Diaye, Aichetou Cheikh Mohamedou Mahmoud, El Benany Mohamed Department of Mathematics and Computer Science University of Nouakchott Nouakchott Mauritania LAROSERI Laboratory Computer Sciences Department Faculty of Sciences University of Chouaib Doukkali El Jadida Morocco

ISBN: (纸本)9783031262531

Today, In Mauritania social media such as Facebook and Tweeter are widely used not only for family or friendly communication but also for sharing, advertising, especially used in the field of social policy, therefore a large amount of information produced and reflects the intentions of these users. This data must be used in an optimal way in order to help decision-makers when making decisions. The aim of this survey is to compare the existing methods of analyzing data flows in real time applying to social networks. We therefore seek to compare the strategic BI models of real-time social media data analysis applied to government information in Mauritania. This model will allow to give on the one hand a clear and timely vision, by grouping the topical subjects by their importance according to the audiences of the public, also indicating their feelings towards the publications by classifying them according to their polarity. On the other hand, only relevant data can be saved in the data warehouse which will then be searchable by government sectors with the possibility of accessing only the data that concerns it, which will lead to the speed of BI analysis process. © 2023, The Author(s), under exclusive license to Springer Nature Switzerland AG.

关键词： Sentiment analysis

来源：评论

学校读者我要写书评

暂无评论

Binary Stochastic Flip Optimization for Training Binary Neural Networks

Binary Stochastic Flip Optimization for Training Binary Neur...

引用

2025 IEEE International Conference on Acoustics, Speech, and Signal Processing, ICASSP 2025

作者： Shibuya, Tatsukichi Inoue, Nakamasa Kawakami, Rei Sato, Ikuro Department of Computer Science Institute of Science Tokyo Tokyo Japan Department of Systems and Control Engineering Institute of Science Tokyo Tokyo Japan Denso IT Laboratory Tokyo Japan

ISBN: (纸本)9798350368741

For deploying deep neural networks on edge devices with limited resources, binary neural networks (BNNs) have attracted significant attention, due to their computational and memory efficiency. However, once a neural network is binarized, finetuning it on edge devices becomes challenging because most conventional training algorithms for BNNs are designed for use on centralized servers and require storing real-valued parameters during training. To address this limitation, this paper introduces binary stochastic flip optimization (BinSFO), a novel training algorithm for BNNs. BinSFO employs a parameter update rule based on Boolean operations, eliminating the need to store real-valued parameters and thereby reducing memory requirements and computational overhead. In experiments, we demonstrated the effectiveness and memory efficiency of BinSFO in finetuning scenarios on six image classification datasets. BinSFO performed comparably to conventional training algorithms with a 70.7% smaller memory requirement. Code is released at https://***/TatsukichiShibuya/ICASSP2025_BinSFO. © 2025 IEEE.

关键词：

来源：评论

学校读者我要写书评

暂无评论

Efficient graph representation framework for chemical molecule similarity tasks 6

Efficient graph representation framework for chemical molecu...

引用

6th IEEE International Conference on Artificial Intelligence and Knowledge Engineering, AIKE 2023

作者： Ma, Jiaji Lim, Seung-Hwan Department of Computer Science University of Virginia CharlottesvilleVA United States Computer Science and Mathematics Division Oak Ridge National Laboratory Oak RidgeTN United States

ISBN: (纸本)9798350331288

Graph data has emerged in numerous scientific domains and machine learning techniques have been widely used for analysis and learning of diverse data for prediction and decision. Machine learning techniques can readily address complex problems by leveraging their structural information. But graphs cannot be directly used for existing machine learning algorithms unless encoded as vectors. The problem of efficient representation of graphs is a substantial challenge in graph machine learning. In this paper, we propose a novel two-stage framework for the representation of chemical molecule graphs based on the strengths of Graph Isomorphism Networks (GINs) and Siamese autoencoders. In the first stage, the GIN model is constructed and trained using the structural information of chemical molecule graphs. Node attributes, edge attributes, and edge indices are used as input data, while graph attributes are used as labels. The GIN model effectively captures the structural characteristics of graphs and can accurately predict graph attributes, i.e., molecular properties. It also generates Graph Embeddings, represented as vectors that encode the structural information of graphs. In the second stage, Graph Embedding vectors are further optimized for downstream similarity tasks while preserving the graph structural information. The Siamese autoencoder is constructed and trained, which reduces the dimensionality of the Graph Embedding vectors, while maximizing the preservation of structural information in the original high-dimensional vectors. The resulting low-dimensional Graph Embeddings can be effectively utilized for tasks such as approximate nearest neighbor search. The experimental results demonstrate the effectiveness of our proposed framework in accurately predicting graph similarity. © 2023 IEEE.

关键词： Vectors

来源：评论

学校读者我要写书评

暂无评论

QCrypt: Advanced Quantum-based Image Encryption for Secure Satellite Data Transmission 3

QCrypt: Advanced Quantum-based Image Encryption for Secure S...

引用

3rd International Conference on Electrical, Electronics, Information and Communication Technologies, ICEEICT 2024

作者： Mohamed Ihsan, P.M. Rajan, Arun Amaithi Vetriselvi, V. Gautham Kumar, G. Praveen Kumar, R. Anna University College of Engineering Guindy Security Research Laboratory Department of Computer Science and Engineering Tamil Nadu Chennai India

ISBN: (纸本)9798350369083

With the advancements in satellite communication, there arises a demand for the secure transmission of satellite and remote sensing images to the ground monitoring stations which has significant challenges due to the increased threat of cyber-attacks and the sensitive nature of those images. Existing classical cryptography methods fail to meet the advanced security demands. Thus, there arises a need for advanced cryptography techniques which ensure secure transmission of data in satellite communication. Recent advancements in quantum computing and research have shown that leveraging quantum mechanics and principles enhances security. We propose a novel quantum-based image encryption technique for securing satellite and remote sensing images during transmission named QCrypt. QCrypt utilizes the power of quantum chaotic maps, classic chaotic maps, and DNA encoding techniques to ensure the security of the images. QCrypt has been tested using a remote sensing image dataset and achieved increased performance in terms of resistance to histogram analysis, differential attacks, and chosen- plaintext attacks compared to traditional encryption methods. Our approach addresses the critical need for secure satellite data transmission and the use of quantum-based encryption to meet security demands. © 2024 IEEE.

关键词： Satellite communication systems

来源：评论

学校读者我要写书评

暂无评论

Optimizing Different Methods of Fabricating 3D Printed UV Blocking Sheets for Solar Cells 1

Optimizing Different Methods of Fabricating 3D Printed UV Bl...

引用

1st Renewable Energy Technologies and Modern Communications Systems: Future and Challenges Conference, RETMCS 2024

作者： Alhindi, Amjad Fawzi Elatab, Nazek Baawad, Lujain Khalid Bahabry, Rabab R. Allam, Fahad Laboratory Electrical and Computer Engineering Computer Electrical Mathematical Science and Engineering Thuwal Saudi Arabia College of Science Department of Physical Sciences Jeddah21589 Saudi Arabia Materials Science and Engineering Department King Fahd University of Petroleum and Minerals Dhahran31261 Saudi Arabia

ISBN: (纸本)9798331519445

This study investigates the fabrication of innovative UV-blocking sheets that effectively transmit visible light while simultaneously obstructing harmful ultraviolet (UV) radiation, utilizing Cerium Oxide (CeO2) and Zinc Oxide (ZnO) nanoparticles as UV-blocking agents. Three distinct fabrication methods were explored: The Powder Method, where nanoparticles were incorporated directly into polydimethylsiloxane (PDMS);the Ball milling Mixture method, which involved mixing nanoparticles through ball milling;and the Solvent-Based Mixture method, where nanoparticles were dissolved, mixed, and dried;before incorporation into PDMS. Samples were produced using a 3D printing technique, which is Direct Ink Writing (DIW) and characterized for their optical properties via UV-Vis spectroscopy. The results indicated CeO2 demonstrated superior UV-blocking capabilities (a maximum blockage of ~99%) compared to ZnO, while maintaining high transmittance for visible light (a maximum of ~90%). Notably, the solvent-based method exhibited better results than the ball milling method;however, the powder method had the best results. Therefore, this investigation offers a promising solution for solar panel protection that aligns with Saudi Arabia's Vision 2030 energy objectives. The relentless exposure of devices, particularly solar cells, to harmful UV radiation poses a formidable challenge and leads to annual degradation rates of between −0.8% and −4.9%, besides other factors. These findings underscore the critical importance of developing effective UV-blocking solutions. ©2024 IEEE.

关键词： Polydimethylsiloxane

来源：评论

学校读者我要写书评

暂无评论

Handwriting Recognition Using HOG and Gabor Features 1

引用

13th International Conference on Information Systems and Advanced Technologies, ICISAT 2023

作者： Aouine, Mohammed Gattal, Abdeljalil Abbas, Faycel LIMPAF Laboratory Department of Computer Science University of Bouira Bouira Algeria Echahid Cheikh Larbi Tebessi University Tebessa Algeria

ISBN: (数字)9783031605949

ISBN: (纸本)9783031605932

This work describes an efficient method for offline handwriting, Arabic word recognition, and digit recognition. The system employs a global recognition method, wherein segmentation of words is not used and the image is identified as a whole. To create this system, a preprocessing operation is applied to images of words or digits, which includes size normalization and the removal of diacritical marks and white spaces in columns and rows. The words or digits to be identified were subjected to feature extraction using the Gabor filter and histogram of oriented gradients (HOG). The KNN classifier used these features as descriptors. The proposed technique was evaluated utilizing both the IFN/ENIT databases and the standard MNIST database, and positive outcomes were attained. © The Author(s), under exclusive license to Springer Nature Switzerland AG 2024.

关键词： Gabor filters

来源：评论

学校读者我要写书评

暂无评论

Performance Analysis of a Non-markovian Queue with Differentiated Vacations

引用

SN computer science 2023年第5期4卷 512页

作者： Ait Braham, Karin Taleb, Samira Aissani, Amar Department of Probability and Statistics (RIIMA Laboratory) University of Science and Technology Houari Boumediene (USTHB) Algiers Algeria Department of Computer Science (RIIMA Laboratory) University of Science and Technology Houari Boumediene (USTHB) Algiers Algeria

Queueing systems with vacations have been used in various applications, such as cellular networks and sensor networks, where base station sleeping is an effective way to improve the energy-efficiency of the network (it corresponds to a power-saving mode). In this paper, we consider a vacation queue model with Poisson arrivals, general service distribution, and differentiated vacations. The server takes Type I vacation of random duration after serving all customers in the system. At the end of this vacation, if the system is still empty, the server takes a new random vacation of Type II. When returning from a Type I or II vacation, if there are some customers in the system, the server immediately starts serving customers until the system becomes empty again. Type II vacations are repeated as long as the system is empty upon the completion of a vacation. We give the necessary and sufficient condition for the system to be stable and obtain the joint distribution of the server state and the number of customers in the system in terms of generating functions. Some performance measures are derived. The asymptotic behavior of the system length is examined under heavy traffic. We are also interested in the sojourn time and the server busy period. The effect of the system parameters on the performance measures is shown through an illustrative numerical example. We particularly pay attention to the power consumption by considering the vacation as the power-saving mode. © 2023, The Author(s), under exclusive licence to Springer Nature Singapore Pte Ltd.

关键词： Differentiated vacations General distribution of service Generating function Laplace transform Performance measures Vacation queues

来源：评论

学校读者我要写书评

暂无评论

Belief Combination of Classifiers for Incomplete Data

引用

IEEE/CAA Journal of Automatica Sinica 2022年第4期9卷 652-667页

作者： Zuowei Zhang Songtao Ye Yiru Zhang Weiping Ding Hao Wang IEEE the Research Center for Optical Fiber Sensing Zhejiang Laboratory the School of Automation Northwestern Polytechnical University the School of Cyber Engineering Xidian University the Department of Computer Science CY Cergy Paris Université the School of Information Science and Technology Nantong University the School with of Cyber Engineering Xidian University the Department of Computer Science Norwegian University of Science and Technology

Data with missing values,or incomplete information,brings some challenges to the development of classification,as the incompleteness may significantly affect the performance of *** this paper,we handle missing values in both training and test sets with uncertainty and imprecision reasoning by proposing a new belief combination of classifier(BCC)method based on the evidence *** proposed BCC method aims to improve the classification performance of incomplete data by characterizing the uncertainty and imprecision brought by *** BCC,different attributes are regarded as independent sources,and the collection of each attribute is considered as a ***,multiple classifiers are trained with each subset independently and allow each observed attribute to provide a sub-classification result for the query ***,these sub-classification results with different weights(discounting factors)are used to provide supplementary information to jointly determine the final classes of query *** weights consist of two aspects:global and *** global weight calculated by an optimization function is employed to represent the reliability of each classifier,and the local weight obtained by mining attribute distribution characteristics is used to quantify the importance of observed attributes to the pattern *** comparative experiments including seven methods on twelve datasets are executed,demonstrating the out-performance of BCC over all baseline methods in terms of accuracy,precision,recall,F1 measure,with pertinent computational costs.

关键词： Classifier fusion classification evidence theory incomplete data missing values

来源：评论

学校读者我要写书评

暂无评论

MEC Implementation in NFV Architecture Supporting 5G End-to-End Network Slicing 2

MEC Implementation in NFV Architecture Supporting 5G End-to-...

引用

2nd International Conference on 6G Networking, 6GNet 2023

作者： Chan, Yao-Chia Yen, Li-Hsing Wang, Tse-Han Tseng, Chien-Chao College of Computer Science National Yang Ming Chiao Tung University Department of Computer Science Hsinchu Taiwan Chunghwa Telecom Laboratories Network Management Laboratory Taoyuan Taiwan

ISBN: (纸本)9798350306736

Multi-access Edge Computing (MEC) utilizes computing platforms in the proximity of end users to provide cloud-based services with low latency and high reliability. The 3rd Generation Partnership Project (3GPP) introduces network slicing to provide diverse network services in 5G networks and has been working with European Telecommunications Standards Institute (ETSI) to standardize the integration of MEC with 5G to support network slicing. However, the realization of network slicing in MEC is still in early stage. All the existing studies on network slicing in MEC were based on 4G Evolved Packet Core (EPC), which do not conform to the network slicing standard developed by 3GPP. This work extends our prior work on management and orchestration (MANO) framework for 5G end-to-end (E2E) network slicing by incorporating the support for MEC. Our contribution is twofold. We report our MEC implementation in NFV architecture using all open-source software. We also make the MEC support network slicing and integrate it into an E2E networking slicing framework. The experimental results show that our MANO framework efficiently deploys k8s clusters, MEC platforms, and E2E network slices with short deployment time and low CPU and memory usage. Furthermore, the round-trip time to access services provided by MEC is much shorter than that of core networks. © 2023 IEEE.

关键词： 5G mobile communication systems

来源：评论

学校读者我要写书评

暂无评论

RelayRec: Empowering Privacy-Preserving CTR Prediction via Cloud-Device Relay Learning 23

RelayRec: Empowering Privacy-Preserving CTR Prediction via C...

引用

23rd ACM/IEEE International Conference on Information Processing in Sensor Networks, IPSN 2024

作者： Deng, Yongheng Wang, Guanbo Yue, Sheng Rao, Wei Zu, Qin Wang, Wenjie Chen, Shuai Ren, Ju Zhang, Yaoxue Meituan Beijing China Tsinghua University BNRist Department of Computer Science and Technology Beijing China Zhongguancun Laboratory Beijing China

ISBN: (纸本)9798350362015

Click-through rate (CTR) prediction holds paramount importance across numerous applications, profoundly impacting user experience and business profitability. The freshness of a CTR prediction model significantly influences its performance, since users' needs and interests may be changing over time, thereby requiring the model to be updated frequently. However, stringent data protection regulations have constrained the collection of users' personal data, posing challenges to traditional model refreshing strategies that rely on centralized data collection. On-device learning techniques, such as federated learning (FL), offer a viable solution by enabling model training on devices without compromising user privacy. Nevertheless, the scarcity of training data with diverse distributions among devices presents considerable obstacles to on-device learning effectiveness. To address these challenges, we introduce RelayRec, a cloud-device relay learning framework designed for privacy-preserving CTR prediction. To establish competent initial models for devices, RelayRec categorizes pre-regulation cloud data into user preference groups, training preference-specific models for devices. Furthermore, a cloud-based automated model selector is developed to identify suitable initial models for devices. To elevate the relay learning performance of these initial models, we incorporate a personalized collaborative learning mechanism that aggregates device models based on user preferences. Extensive experimental evaluations underscore RelayRec's superior performance compared to state-of-the-art benchmarks, affirming its efficacy in privacy-preserving CTR prediction. © 2024 IEEE.

关键词： Learning systems

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：