检索结果-内蒙古大学图书馆

arXiv 2025年

作者： Felemban, Muhamad Almutairi, Abdulrahman Ghafoor, Arif Computer Engineering Department The Information and Computer Science Department The Interdisciplinary Research Center for Intelligent Secure Systems KFUPM Saudi Arabia Purdue University West LafayetteIN United States Computer Engineering Department The College of Computer and Information Science in King Saud University Riyadh Saudi Arabia

Organizations are increasingly moving towards the cloud computing paradigm, in which an on-demand access to a pool of shared configurable resources is provided. However, security challenges, which are particularly exacerbated by the multitenancy and virtualization features of cloud computing, present a major obstacle. In particular, sharing of resources among potentially untrusted tenants in access controlled cloud datacenters can result in increased risk of data leakage. To address such risk, we propose an efficient risk-aware sensitive property-driven virtual resource assignment mechanism for cloud datacenters. We have used two information-theoretic measures, i.e., KL-divergence and mutual information, to represent sensitive properties in the dataset. Based on the vulnerabilities of cloud architecture and the sensitive property profile, we have formulated the problem as a cost-drive optimization problem. The problem is shown to be NP-complete. Accordingly, we have proposed two heuristics and presented simulation based performance results for cloud datacenters with multiple sensitivity. © 2025, CC BY.

关键词： Cloud platforms

来源：评论

学校读者我要写书评

暂无评论

Dynamic Gesture Recognition for Sign Language Using Long Short-Term Memory Networks

Dynamic Gesture Recognition for Sign Language Using Long Sho...

引用

Disruptive Technologies (ICDT), International Conference on

作者： Rohan Singh Rahul Sharma Sushil Kumar Gautam Youddha Beer Singh Computer Science and Engineering Department GCET Greater Noida UP India

ISBN: (数字)9798331519582

ISBN: (纸本)9798331519599

A significant defect in our society is the social divide between people with disabilities and those without. Communication is one of the most significant characteristics of humans, who are thought of as social animals. One of the biggest challenges for those with hearing and vocal impairments is communication. For someone who has hearing and voice impairments, this incapacity to communicate causes regular issues and interferes with their everyday tasks. We have suggested a way to get beyond this communication obstacle in our research work. Everyone can use this solution with ease, and with a few tweaks, it can be made to function on the majority of systems having camera modules. Our method employs an integrated camera module to record hand movements in real time, based on landmarks or hand key points. Proposed LSTM based model have been evaluated in real time environment and a significant result have been reported as mentioned in the result section.

关键词： Hands Deep learning Sign language Image processing Auditory system Cameras People with disabilities Real-time systems Disruptive technologies Long short term memory

来源：评论

学校读者我要写书评

暂无评论

TCP-Cubiwood: Enhanced Cubic Congestion Control for mmWave and Future Networks

TCP-Cubiwood: Enhanced Cubic Congestion Control for mmWave a...

引用

International Communication Systems and Networks and Workshops, COMSNETS

作者： Ujjayant Prakash Anand M. Baswade Department of Computer Science and Engineering Indian Institute of Technology Bhilai India

ISBN: (数字)9798331531195

ISBN: (纸本)9798331531201

The rise in mobile internet usage especially using cellular networks demands efficient performance for web traffic, primarily made up of short TCP flows. For TCP, Cubic is the most widely deployed congestion control algorithm. mmWave and THz offer very high bandwidth and low latency, enabling faster speeds and reduced delay for companies deploying innovative applications, positioning them as the future of wireless technology. As such, we are compelled to thoroughly evaluate the performance of Cubic in various mmWave specific NLoS/LoS scenarios. In the process, we find Cubic really struggling when it comes to static NLoS scenarios and long connections particularly because Cubic keeps on attempting recovery from slow-start but fails due to the NLoS condition. This is where TCP Cubiwood comes to the rescue with an improvement of 189%in a simulated static NLoS scenario with results comparable to Cubic in wired conditions as well, thus becoming a suitable contender for replacing the historical algorithm. Delayed Congestion Response (DCR) tries to make TCP more robust towards channel errors in wireless networks so we also propose an additional flavor of Cubiwood, called Cubiwood-DCR which increases performance slightly in both wired and wireless high frequency scenarios. We extensively evaluate Cubiwood against Cubic, both with and without DCR enabled and make our code available for reproducibility and further research in the same direction. Moreover, the simplicity of this method makes our proposal a much more attractive solution. We believe the proposed solution is easy to implement, can replace Cubic and has the potential to improve performance in THz communication as well which is one of the competing spectrum bands for 6G networks.

关键词： Wireless sensor networks Codes Wireless networks Sockets Throughput Topology High frequency Millimeter wave communication Standards Terahertz communications

来源：评论

学校读者我要写书评

暂无评论

Optimizing Bengali Sentiment Analysis: A Comparison of Countvectorizer and TF-IDF with Machine Learning

Optimizing Bengali Sentiment Analysis: A Comparison of Count...

引用

International Conference on Electrical, computer and Communication engineering (ECCE)

作者： Umme Ayman Anzir Rahman Khan Mehraj Hossain Mahi Tamanna Akter Zannatul Mawa Koli Md. Hasan Imam Bijoy Department of Computer Science and Engineering Daffodil International University Dhaka Bangladesh

ISBN: (数字)9798350357509

ISBN: (纸本)9798350357516

In the realm of NLP, sentiment analysis is an indicator for identifying emotions in human language. Compared to the English language, research on low-resource languages like Bengali has not reached its full potential yet. This study evaluates the performance of CountVectorizer and Tf-Idf vectorizers to analyze sentiment in Bengali text accurately. Here we have applied SVM, MNB, RF and Xgb to the self-generated Bengali text dataset to evaluate the effectiveness of these feature extraction techniques. These techniques transform Bengali text into numerical representations for precise sentiment classification, addressing challenges like morphological complexity and syntactic nuances, highlighting the need for accurate representation of the text's sentimental features. Before applying machine learning algorithms, the dataset is preprocessed by utilizing several text preprocessing techniques to enhance its performance. We have obtained the highest accuracy of 91.15% with the MNB using the CountVectorizer strategy, outperforming the Tf-Idf slightly, which reached the accuracy of 91.00% is obtained with the MNB. The overall accuracy with CountVectorizer is 85.38%-91.15% and with Tf-Idf is 85.34% -91.00%. It is also observed that CountVectorizer has performed better than Tf-Idf comparatively. Eventually, the findings highlight the strengths and limitations of each technique, which are imperative for future research and study in NLP for Bengali. This work advances Bengali sentiment analysis and improves sentiment analysis's accuracy by determining the most effective feature extraction method.

关键词： Deep learning Sentiment analysis Analytical models Accuracy Machine learning algorithms Computational modeling Transforms Linguistics Feature extraction Complexity theory

来源：评论

学校读者我要写书评

暂无评论

LLMPi: Optimizing LLMs for High-Throughput on Raspberry Pi

arXiv

引用

arXiv 2025年

作者： Ardakani, Mahsa Malekar, Jinendra Zand, Ramtin Department of Computer Science and Engineering University of South Carolina United States

Deploying Large Language Models (LLMs) on resource-constrained edge devices like the Raspberry Pi presents challenges in computational efficiency, power consumption, and response latency. This paper explores quantization-based optimization techniques to enable high-throughput, energy-efficient execution of LLMs on low-power embedded systems. Our approach leverages k-quantization, a Post-Training Quantization (PTQ) method designed for different bit-widths, enabling efficient 2-bit, 4-bit, 6-bit, and 8-bit weight quantization. Additionally, we employ ternary quantization using Quantization-Aware Training (QAT) for BitNet models, allowing for more effective adaptation to lower-bit representations while preserving accuracy. Our findings highlight the potential of quantized LLMs for real-time conversational AI on edge devices, paving the way for low-power, high-efficiency AI deployment in mobile and embedded applications. This study demonstrates that aggressive quantization strategies can significantly reduce energy consumption while maintaining inference quality, making LLMs practical for resource-limited environments. Copyright © 2025, The Authors. All rights reserved.

关键词： Embedded systems

来源：评论

学校读者我要写书评

暂无评论

TLBEMSE: design of a transfer learning-based bioinspired ensemble model for preemptive detection of stress and emotional disorders

引用

Neural Computing and Applications 2025年 1-26页

作者： Hole, Komal Rajendra Anand, Divya Mohanty, Sachi Nandan Rathore, Rajkumar Singh lvarez, Roberto Marcelo School of Computer Science and Engineering Lovely Professional University Punjab Phagwara 144411 India School of Computer Science and Engineering VIT-AP University Andhra Pradesh Amravati 522237 India Department of Computer Science Cardiff School of Technologies Cardiff Metropolitan University Llandaff CampusWestern Avenue Cardiff CF5 2YB United Kingdom Higher Polytechnic School Universidad Europea del Atlántico C/Isabel Torres 21 Santander 39011 Spain Department of Engineering Universidad Internacional Iberoamericana Arecibo Puerto Rico 00613 United States Department of Project Management Universidad Internacional Iberoamericana Campeche 24560 Mexico Prof. Ram Meghe Institute of Technology & Research Badnera Maharashtra Amravati 444701 India

Electroencephalogram signals are used to depict emotional and stress disorders. To overcome issues of existing models, novel transfer learning-based bioinspired ensemble model for preemptive detection of stress and emotional disorders is discussed. The proposed model includes features of mel-frequency cepstral coefficient, iVector, cosine, Fourier and wavelet components. A combination of these features is processed via gray wolf optimization which aims at variance maximization across. The selected features are converted into 2D representation and processed via a transfer learning-based convolutional neural network model combining ResNet 101, MobileVNet, and YoLo models. The classified results from these models are further cross-validated via use of ensemble classification that combines Naïve Bayes, support vector machine, random forest, logistic regression, and multilayer perceptron models. These classifiers perform several post-processing tasks involving identification of disease spread probability, estimation of future diseases, etc. The proposed model was trained on DEAP and interface datasets, compared w.r.t. various state-of-the-art methods, in relation to accuracy, recall, precision, area under the curve, and delay performance. Based on this performance, proposed model’s effectiveness was noticed, showcasing 8.5% higher accuracy, 8.3% higher precision, 5.9% better recall, 4.5% better AUC, and 14.9% faster classification performance, which makes it highly useful for clinical deployments. © The Author(s), under exclusive licence to Springer-Verlag London Ltd., part of Springer Nature 2025.

关键词： Bioinspired Electroencephalogram Ensemble GWO Stress

来源：评论

学校读者我要写书评

暂无评论

Colorization Quality Assessment with CLIP

Colorization Quality Assessment with CLIP

引用

IEEE International Conference on Consumer Electronics (ICCE)

作者： Shunta Shimizu Hiroshi Ishikawa Department of Computer Science and Communications Engineering Waseda University Tokyo Japan

ISBN: (数字)9798331521165

ISBN: (纸本)9798331521172

Automatic colorization of grayscale images defies simple evaluation by quantitative metrics because there can be multiple equally good but different “truths,” making consistent evaluation challenging. To address this problem, we propose an evaluation model for image colorization that utilizes prompt learning, with novel components Color-Query Cross Attention Block and addition of Positive and Negative input images, based on the prior knowledge CLIP possesses. This complements conventional metrics by evaluating naturalness of the colorization based on the scene content.

关键词： Measurement Image quality Evaluation models Gray-scale Quality assessment Reliability Consumer electronics

来源：评论

学校读者我要写书评

暂无评论

Traditional Art and Craft Item Recognition Using Deep Learning

Traditional Art and Craft Item Recognition Using Deep Learni...

引用

International Conference on Electrical, computer and Communication engineering (ECCE)

作者： Tawhid Ahmed Komol Rinvi Jaman Riti Pulak Deb Nath Shrabon Datta Shayla Sharmin Sheak Rashed Haider Noori Department of Computer Science and Engineering Daffodil International University Dhaka Bangladesh

ISBN: (数字)9798350357509

ISBN: (纸本)9798350357516

Recognizing and preserving traditional art and craft is essential to save cultural heritage and gain global acceptance. To promote our culture and tradition worldwide, this research uses deep learning methods to classify traditional art and craft products from Bengal. We scraped 1,872 photos from online and divided into nine categories to create the dataset. Our methodology includes 80:20 train-test, data labelling, and image augmentation. Apart from that we trained InceptionResNetV2, Xception and VGG16. The Inception-ResNetV2 outperformed all other models with an accuracy of 95.73%. Model performance was measureable via metrics like confusion matrices, classification report and train & validation accuracy loss curve. The results demonstrate the impressive adaptive capabilities of transfer learning models spotting the items of traditional art and craft using InceptionResNetV2.

关键词： Deep learning Adaptation models Art Accuracy Image recognition Transfer learning Loss measurement Cultural differences Global communication Image classification

来源：评论

学校读者我要写书评

暂无评论

A Genetic Algorithm-Based Parameter Selection for Communication-Efficient Federated Learning 28th

A Genetic Algorithm-Based Parameter Selection for Communica...

引用

28th European Conference on Applications of Evolutionary Computation, EvoApplications 2025, held as part of EvoStar 2025

作者： Hassan, Mir Yildirim, Kasim Sinan Iacca, Giovanni Department of Information Engineering and Computer Science University of Trento Trento Italy

ISBN: (纸本)9783031900648

Federated Learning (FL) enables decentralized model training without centralized data collection, but high communication overhead remains a key challenge, particularly in bandwidth-constrained environments like IoT and edge networks. Existing FL methods transmit full model updates, which leads to high communication costs. In this paper, we propose Genetic Algorithm-based Selective Parameter Updates (GASPU), a novel approach that uses a Genetic Algorithm (GA) to selectively transmit model parameter updates, significantly reducing communication overhead while maintaining competitive accuracy. GASPU optimizes binary masks allowing only the most effective parameters to be sent. We validate the GASPU approach on the HAR and KWS datasets, which are representative of realistic FL settings. While achieving a 66% reduction in communication overhead over 100 communication rounds on HAR (from 0.49 MB to 0.16 MB) and a reduction to (from 1.27 MB to 0.44 MB) on KWS, GASPU maintained competitive accuracy with only 10% drop. Existing methods achieve higher accuracies (above 80%) but at significantly higher communication costs. Further experiments on the MNIST benchmarking dataset confirm GASPU ’s generalizability, achieving only a 0.25% drop in accuracy and a 52% reduction in communication overhead over 10 communication rounds. © The Author(s), under exclusive license to Springer Nature Switzerland AG 2025.

关键词： Genetic algorithms

来源：评论

学校读者我要写书评

暂无评论

Enhancing Personalization: A Machine Learning Approach to Modern Recommender Systems

Enhancing Personalization: A Machine Learning Approach to Mo...

引用

Disruptive Technologies (ICDT), International Conference on

作者： Supriya Saxena Bharat Bhushan Department of Computer Science and Engineering Sharda University Greater Noida India

ISBN: (数字)9798331519582

ISBN: (纸本)9798331519599

In the digital era where Industrie 4.0 has emerged, most of the application are either web based or mobile based and also, people are becoming very specific in selecting their products and amenities over internet. This preference in choices of users makes it difficult for the service provider to understand which product or service they should provide to the user so that the user can select appropriately and that too in less time. This has led to the use of recommendation system for these purposes. Also, the recommendation system helps manufacturers to understand that which product has higher consumption and which product has lower consumption. This paper tries its best to first explain the foundation concept of recommendation system and its forms. Also, the paper tries its best to deep dive into all the relevant research that has been carried out to understand the merits as well as limitation of if recommendation system. This work deals with various machine learning algorithms such as Neural Collaborative Filtering (NCF), Embedding Layers, Deep Neural Networks (DNN), Matrix Factorization, Hybrid Model: Collaborative Filtering (SVD), Content-Based Filtering. The paper also highlights the vital applications of recommendation system that are useful in real life.

关键词： Industries Machine learning algorithms Accuracy Filtering Collaborative filtering Machine learning Artificial neural networks Electronic commerce Disruptive technologies Recommender 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：