检索结果-内蒙古大学图书馆

8th International Workshop and Tutorial on Interactive Adaptive Learning, IAL@ECML-PKDD 2024

作者： Röder, Manuel Schleif, Frank-Michael Faculty of Computer Science and Business Information Systems TUAS Würzburg-Schweinfurt Würzburg Germany Faculty of Technology Bielefeld University Bielefeld Germany Center for Artificial Intelligence and Robotics Würzburg Germany

This extended abstract explores the integration of federated learning with deep transfer hashing for distributed prediction tasks, emphasizing resource-efficient client training from evolving data streams. Federated learning allows multiple clients to collaboratively train a shared model while maintaining data privacy - by incorporating deep transfer hashing, high-dimensional data can be converted into compact hash codes, reducing data transmission size and network loads. The proposed framework utilizes transfer learning, pre-training deep neural networks on a central server, and fine-tuning on clients to enhance model accuracy and adaptability. A selective hash code sharing mechanism using a privacy-preserving global memory bank further supports client fine-tuning. This approach addresses challenges in previous research by improving computational efficiency and scalability. Practical applications include Car2X event predictions, where a shared model is collectively trained to recognize traffic patterns, aiding in tasks such as traffic density assessment and accident detection. The research aims to develop a robust framework that combines federated learning, deep transfer hashing and transfer learning for efficient and secure downstream task execution. © 2024 Copyright for this paper by its authors.

关键词： Federated learning

来源：评论

学校读者我要写书评

暂无评论

Toward Utilizing Bidirectional Multi-head Attention Technique for Automatic Correction of Grammatical Errors

IAENG International Journal of Computer Science

引用

IAENG International Journal of computer science 2024年第11期51卷 1886-1897页

作者： Mahmoud, Zeinab Kryvinska, Natalia Abdalsalm, Mohammed Solyman, Aiman Alfatemi, Ali Musyafa, Ahmad School of Computer Science and Technology Wuhan University of Technology China Department of Information Management and Business Systems Comenius University Slovakia Department of Computer Science University of Milan Italy CIS Department Fordham University BronxNY10458 United States Department of Informatics Engineering Pamulang University South Tangerang15310 Indonesia

Automatic Grammar Error Correction (GEC) models identify and correct a wide range of grammatical errors. Various strategies have been proposed for GEC, with the Neural Machine Translation (NMT) approach being the most effective. However, NMT-based GEC models with encoder-decoder layers rely heavily on the highest layer, leading to potential inaccuracies. Additionally, during inference, exposure bias can cause the model to substitute previously targeted words with incorrect alternatives. Another challenge is data scarcity. This paper introduces a GEC model leveraging the seq-to-seq Transformer framework, specifically designed for low-resource languages like Arabic. We propose a method to generate noise in the text to create synthetic parallel data, addressing the data constraints. Inspired by Capsule Networks (CapsNet), we incorporate CapsNet in GEC to dynamically aggregate information from multiple layers. In order to mitigate exposure bias, we incorporated a bidirectional training approach and a regularization term using Kullback-Leibler divergence to align left-to-right and right-to-left models. Experiments on two benchmarks demonstrate that our model outperforms current Arabic GEC models, achieving the highest scores. The code is available on GitHub (https://***/Zainabobied/ArabicGEC). © (2024), (International Association of Engineers). All rights reserved.

关键词： Neural machine translation

来源：评论

学校读者我要写书评

暂无评论

Boosting speech recognition performance: a robust and accurate ensemble method based on HMMs

引用

International Journal of Intelligent systems Technologies and Applications 2024年第1期22卷 41-76页

作者： Hazmoune, Samira Bougamouza, Fateh Mazouzi, Smaine Benmohammed, Mohamed Department of Computer Science Faculty of Sciences University of 20 Août 1955-Skikda Skikda Algeria Department of Software Technologies and Information Systems Faculty of New Technologies of Information and Communication University Constantine 2 Constantine Algeria

In this paper, we propose an ensemble method based on hidden Markov models (HMMs) for speech recognition. Our objective is to reduce the impact of the initial setting of training parameters on the final model while improving accuracy and robustness, particularly in speaker independent systems. The main idea is to exploit the sensitivity of HMMs to the initial setting of training parameters, thus creating diversity among the ensemble members. Additionally, we perform an experimental study to investigate the potential relationship between initial training parameters and ten diversity measures from literature. The proposed method is assessed on a standard dataset from the UCI machine-learning repository. Results demonstrate its effectiveness in terms of accuracy and robustness to intra-class variability, surpassing basic classifiers (HMM, KNN, NN, SVM) and some previous works in the literature including those using deep learning algorithms such as convolutional neural networks (CNNs) and long short-term memory (LSTM). Copyright © 2024 Inderscience Enterprises Ltd.

关键词： Hidden Markov models

来源：评论

学校读者我要写书评

暂无评论

IOT Based Smart Parking System Using Ensemble Learning

引用

Intelligent Automation & Soft Computing 2023年第6期36卷 3637-3656页

作者： Walaa H.Elashmawi Ahmad Akram Mohammed Yasser Menna Hisham Manar Mohammed Noha Ihab Ahmed Ali Department of Computer Science Faculty of Computers and InformaticsSuez Canal University4.5 Km the Ring RoadIsmailiaIsmailia41522Egypt Department of Computer Engineering and Software Systems Faculty of EngineeringAin Shams UniversityAl-Abbaseya4543070Egypt Department of Computer Science Faculty of Computer ScienceMisr International University28 KM Cairo–Ismailia RoadCairo44971Egypt College of Computer Engineering and Sciences Prince Sattam Bin Abdulaziz UniversityAl-Kharj11942Saudi Arabia Higher Future Institute for Specialized Technological Studies Cairo3044Egypt

Parking space is usually very limited in major cities,especially Cairo,leading to traffic congestion,air pollution,and driver *** car parking systems tend to tackle parking issues in a non-digitized *** systems require the drivers to search for an empty parking space with no guaran-tee of finding any wasting time,resources,and causing unnecessary *** address these issues,this paper proposes a digitized parking system with a proof-of-concept implementation that combines multiple technological concepts into one solution with the advantages of using IoT for real-time tracking of park-ing *** authentication and automated payments are handled using a quick response(QR)code on entry and *** experiments were done on real data collected for six different locations in Cairo via a live popular times *** machine learning models were investigated in order to estimate the occu-pancy rate of certain ***,a clear analysis of the differences in per-formance is illustrated with the final model deployed being *** has achieved the most efficient results with a R^(2) score of 85.7%.

关键词： IoT XGBoost linear regression random forest ensemble learning isolation forest

来源：评论

学校读者我要写书评

暂无评论

Systematic Literature Review on Aspect-Based Sentiment Analysis: Insights from Community Reports

Systematic Literature Review on Aspect-Based Sentiment Analy...

引用

2024 International of Seminar on Application for Technology of Information and Communication, iSemantic 2024

作者： Kamal, Muhammad Rikzam Affandy Pujiono Hasibuan, Zainal Arifin Dian Nuswantoro University Faculty of Computer Science Semarang Indonesia Uin Kh Abdurrahman Wahid Faculty of Islamic Economics and Business Pekalongan Indonesia

ISBN: (纸本)9798350366648

This systematic literature review examines the utilization of datasets, methodologies, and evaluation metrics in Aspect-Based Sentiment Analysis (ABSA) across 44 scholarly publications. Our analysis identifies prevailing trends in dataset selection, with Twitter and TripAdvisor being the most frequently used sources due to their rich and diverse content. We categorize the various algorithms, ranging from lexicon-based approaches like VADER to advanced machine learning and deep learning models such as SVM and BERT. Standard evaluation metrics, including accuracy, F1-Score, precision, and recall, are analyzed to understand how performance is assessed. The review highlights a shift towards deep learning and hybrid models, reflecting recent technological advancements. Our findings offer a comprehensive overview of the current state of ABSA research, identifying existing gaps and providing insights for future studies to enhance the effectiveness of sentiment analysis techniques. © 2024 IEEE.

关键词： Contrastive Learning

来源：评论

学校读者我要写书评

暂无评论

Image-Based Sentiment Analysis Using InceptionV3 Transfer Learning Approach

引用

SN computer science 2023年第3期4卷 242页

作者： Meena, Gaurav Mohbey, Krishna Kumar Kumar, Sunil Chawda, Rahul Kumar Gaikwad, Sandeep V. Department of Computer Science Central University of Rajasthan Ajmer India Department of Computer Science Assam University Silchar India School of Business Woxsen University Hyderabad India Faculty of Computer Science and Application SMPICA Charotar University of Science and Technology Gujrat Anand India

There has been a recent shift from using text-based sentiment analysis in favor of an image-based method. In recent years, transfer learning methods have been widely utilized in developing a comprehensive image sentiment analysis approach. Deep learning algorithms have produced remarkable outcomes in a variety of contexts. Image-based sentiment analysis presents many difficulties, but there also appears to be much space for development. A significant improvement over prior work is provided by an InceptionV3 approach that can easily focus on huge body portions like a human face. This research improves image categorization performance with InceptionV3, a popular deep convolutional neural network, and other deep features. Using a Convolutional Neural Network based on InceptionV3 architecture, we identify and classify emotions using the famous CK + , FER2013, and JAFFE datasets. Experiments reveal that the proposed model achieves 99.5% accuracy on the CK + dataset. Also, the accuracies of JAFFE and FER2013 are 86% and 73%, respectively. A person’s emotional patterns and mental health can be evaluated using this approach. © 2023, The Author(s), under exclusive licence to Springer Nature Singapore Pte Ltd.

关键词： Artificial Intelligence Deep learning Image sentiment analysis InceptionV3 Transfer learning

来源：评论

学校读者我要写书评

暂无评论

Advancements in Multi-View Human Activity Recognition for Ambient Assisted Living

Advancements in Multi-View Human Activity Recognition for Am...

引用

2024 Multimedia University Engineering Conference, MECON 2024

作者： Bari, Ahsanul Farid, Fahmid Al Sarker, Md. Tanjil Mansor, Sarina Karim, Hezerul Abdul Bhuiyan, Md Roman Bannah, Hasanul Multimedia University Faculty of Engineering Cyberjaya63100 Malaysia Berlin School of Business and Informatics Faculty of Computer Science and Informatics Berlin12043 Germany

ISBN: (纸本)9798331530747

This Ambient Assisted Living (AAL) uses technology to improve the well-being, autonomy, and security of seniors and disabled individuals. AAL services depend on Human Activity Recognition to detect human behaviors from sensor data. Compared to single-view models, multi-view HAR, which aggregates data from multiple sensors, provides improved accuracy and reliable detection. In this review, we assess recent progress in multi-view HAR for AAL, focusing on the value of lightweight deep learning models. We cover the development of HAR systems, as well as the emergence of multi-view datasets and highly efficient deep learning models that are explicitly designed for AAL scenarios. Furthermore, we address problems such as data synchronization, data protection, and adaptability, as well as possible remedies like sensor fusion and transfer learning. As a result, we also demonstrate the potential of multi-view HAR to increase AAL services and outlining future research trends in providing adaptive, efficient, and privacy-aware systems. © 2024 IEEE.

关键词： Adversarial machine learning

来源：评论

学校读者我要写书评

暂无评论

Handwritten Alphabet Recognition Using Convolutional Neural Networks

Handwritten Alphabet Recognition Using Convolutional Neural ...

引用

2024 International Conference Automatics and Informatics, ICAI 2024

作者： Jamal, Yasir Neemuchwala, Sakina Juzar Kouatly, Rand Hassan, Ahmed Puthiyapurayil, Nafeesath Parappurath Karunanithi, Meerah Institute of Engg. Sciences & Tech Faculty of Computer Science & Engg. Gik Topi Pakistan University of Europe for Applied Sciences Department of Business Potsdam Germany Berlin School of Business and Innovation Department of Computer Science Berlin Germany

ISBN: (纸本)9798350353907

Instruction of hand-written alphabet recognition using an intelligent adaptive Optical Character Recognition (OCR) system is one of the vital pattern recognition problems, which has a very broad spectrum of frequent applications in various fields, such as language comprehension and archival document investigation. Instruction Recognition has seen a lot of progress in the data recognition field, but the handwritten alphabet recognition follows a separate track and computer with this can be a challenge due to the different styles and shapes of handwriting which are not marked. The project provides room for further work in which the Convolutional Neural Network (CNN) model can be trained to interpret handwritten alphabet characters. In connection with the exact naming of the handwritten images in the A-Z collection, deep neural networks are applied. AI-generated images of the outcome corroborate the effectiveness and accuracy of the constructed CNN algorithm, which precisely identifies handwritten characters of the alphabet with an accuracy above 96-97 percent. Insights from this analysis are pivotal in making this research area a success and CNNs are a powerful tool that can be exploited for this purpose. The presentation seems to enjoy the fact that their results are useful in many areas which includes digitization of historical documents, automatic online form processing, as well as well as customized learning system. © 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

Deep Learning with a Novel Concoction Loss Function for Identification of Ophthalmic Disease

引用

computers, Materials & Continua 2023年第9期76卷 3763-3781页

作者： Sayyid Kamran Hussain Ali Haider Khan Malek Alrashidi Sajid Iqbal Qazi Mudassar Ilyas Kamran Shah Department of Computer Science TIMES InstituteMultan60000Pakistan Department of Software Engineering Faculty of Computer ScienceLahore Garrison UniversityLahore54000Pakistan Department of Computer Science Applied CollegeUniversity of TabukTabukSaudi Arabia Department of Information Systems College of Computer Sciences and Information TechnologyKing Faisal UniversityAl Hofuf31982Saudi Arabia Department of Mechanical Engineering College of EngineeringKing Faisal UniversityAl HofufSaudi Arabia

As ocular computer-aided diagnostic(CAD)tools become more widely accessible,many researchers are developing deep learning(DL)methods to aid in ocular disease(OHD)*** eye diseases like cataracts(CATR),glaucoma(GLU),and age-related macular degeneration(AMD)are the focus of this study,which uses DL to examine their *** imbalance and outliers are widespread in fundus images,which can make it difficult to apply manyDL algorithms to accomplish this analytical *** creation of efficient and reliable DL algorithms is seen to be the key to further enhancing detection *** the analysis of images of the color of the retinal fundus,this study offers a DL model that is combined with a one-of-a-kind concoction loss function(CLF)for the automated identification of *** study presents a combination of focal loss(FL)and correntropy-induced loss functions(CILF)in the proposed DL model to improve the recognition performance of classifiers for biomedical *** is done because of the good generalization and robustness of these two types of losses in addressing complex datasets with class imbalance and *** classification performance of the DL model with our proposed loss function is compared to that of the baseline models using accuracy(ACU),recall(REC),specificity(SPF),Kappa,and area under the receiver operating characteristic curve(AUC)as the evaluation *** testing shows that the method is reliable and efficient.

关键词： Deep learning multi-classification focal loss CNN eye disease

来源：评论

学校读者我要写书评

暂无评论

Performance Framework for Virtual Machine Migration in Cloud Computing

引用

computers, Materials & Continua 2023年第3期74卷 6289-6305页

作者： Tahir Alyas Taher M.Ghazal Badria Sulaiman Alfurhood Munir Ahmad Ossma Ali Thawabeh Khalid Alissa Qaiser Abbas Department of Computer Science Lahore Garrison UniversityLahore54000Pakistan School of Information Technology Skyline University CollegeUniversity City SharjahSharjah1797UAE Center for Cyber Security Faculty of Information Science and TechnologyUniversiti Kebansaan Malaysia(UKM)Bangi43600SelangorMalaysia Department of Computer Sciences College of Computer and Information SciencesPrincess Nourah Bint Abdulrahman UniversitySaudi Arabia School of Computer Science National College of Business Administration&EconomicsLahore54000Pakistan Dean of Student Affairs Skyline University CollegeUniversity City SharjahSharjah1797UAE Networks and Communications Department College of Computer Science and Information TechnologyImam Abdulrahman Bin Faisal UniversityDammam31441Saudi Arabia Faculty of Computer and Information Systems Islamic University Madinah Madinah42351Saudi Arabia

In the cloud environment,the transfer of data from one cloud server to another cloud server is called *** can be delivered in various ways,from one data centre to *** research aims to increase the migration performance of the virtual machine(VM)in the cloud *** allow cloud customers to store essential data and ***,server usage has grown dramatically due to the virtualization of computer systems,resulting in higher data centre power consumption,storage needs,and operating *** VMs on one data centre manage share resources like central processing unit(CPU)cache,network bandwidth,memory,and application ***-cloud,VMmigration addresses the performance degradation due to cloud server configuration,unbalanced traffic load,resource load management,and fault situations during data *** speed is influenced by the size of the VM,the dirty rate of the running application,and the latency ofmigration *** a result,evaluating VM migration performance while considering all of these factors becomes a difficult *** effort of this research is to assess migration problems on *** simulation results in Matlab show that if the VMsize grows,themigration time of VMs and the downtime can be impacted by three orders *** dirty page rate decreases,themigration time and the downtime grow,and the latency time decreases as network bandwidth increases during the migration time and post-migration overhead calculation when the VMtransfer is *** the simulated cases of VMs migration were performed in a fuzzy inference system with performance graphs.

关键词： Latency cloud computing dirty page ratio storage migration performance

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：