检索结果-内蒙古大学图书馆

24th IEEE International Conference on Data Mining, ICDM 2024

作者： Ghosh, Indrajeet Chugh, Garvit Md Faridee, Abu Zaher Roy, Nirmalya University of Maryland Department of Information Systems Baltimore County United States Indian Institute of Technology Department of Computer Science and Engineering Jodhpur India Amazon United States

ISBN: (纸本)9798331506681

Recent advancements in deep learning-based wearable human action recognition (wHAR) have improved the capture and classification of complex motions, but adoption remains limited due to the lack of expert annotations and domain discrepancies from user variations. Limited annotations hinder the model's ability to generalize to out-of-distribution samples. While data augmentation can improve generalizability, unsupervised augmentation techniques must be applied carefully to avoid introducing noise. Unsupervised domain adaptation (UDA) addresses domain discrepancies by aligning conditional distributions with labeled target samples, but vanilla pseudo-labeling can lead to error propagation. To address these challenges, we propose μDAR, a novel joint optimization architecture comprised of three functions: (i) consistency regularizer between augmented samples to improve model classification generalizability, (ii) temporal ensemble for robust pseudo-label generation and (iii) conditional distribution alignment to improve domain generalizability. The temporal ensemble works by aggregating predictions from past epochs to smooth out noisy pseudo-label predictions, which are then used in the conditional distribution alignment module to minimize kernel-based class-wise conditional maximum mean discrepancy (kCMMD) between the source and target feature space to learn a domain invariant embedding. The consistency-regularized augmentations ensure that multiple augmentations of the same sample share the same labels;this results in (a) strong generalization with limited source domain samples and (b) consistent pseudo-label generation in target samples. The novel integration of these three modules in μDAR results in a range of 4-12% average macro-F1 score improvement over six state-of-the-art UDA methods in four benchmark wHAR datasets. © 2024 IEEE.

关键词： Deep learning

来源：评论

学校读者我要写书评

暂无评论

Deep Learning-based Cluster Analysis for Healthcare Data Classification

Deep Learning-based Cluster Analysis for Healthcare Data Cla...

引用

2023 International Conference on Data science, Agents and Artificial Intelligence, ICDSAAI 2023

作者： Gayathri, M. Kavitha, V. Srm Institute of Science and Technology Department of Computer Science and Engineering Chennai Kattankulathur India Srm Institute of Science and Technology Department of Data Science and Business Systems Chennai Kattankulathur India

ISBN: (纸本)9798350348910

The use of cutting-edge technology in the medical field results in the production of massive volumes of data on a daily basis. Various categories of information are applied in the domain of healthcare, including clinic information, medical record data, and genetic information. In addition, real-time monitoring in the medical industry produces enormous volumes of data, and properly interpreting such enormous amounts of data is a significant problem. The evaluation of medical information becomes more required so that suitable drugs may be supplied and issues can be avoided by taking appropriate measures depending on the history of the patient. Automation makes data analysis more effective, but performance might deteriorate when there are problems with data integrity, diversity, or consistency. Automation makes data analysis more effective. In order to handle the management of massive amounts of data, many models that are powered by neural networks have been developed;still researchers are currently attempting to develop a superior model that is more accurate. Hence, fuzzy c denotes the process of clustering, whereas generative adversarial networks are used in this study to clusters and classify medical data, aiming to achieve the maximum degree of efficiency in classification. The experiment will use both the benchmark lung cancer sample and the arrhythmias sample. The proposed model achieves a maximum accuracy of 97.3% for dataset 1 and 98.2% for dataset 2, outperforming other methods such as support vector machine, decision tree, and random forest algorithms. © 2023 IEEE.

关键词： Classification (of information)

来源：评论

学校读者我要写书评

暂无评论

Teamwork Optimization with Deep Learning Based Fall Detection for IoT-Enabled Smart Healthcare System

引用

computers, Materials & Continua 2023年第4期75卷 1353-1369页

作者： Sarah B.Basahel Saleh Bajaba Mohammad Yamin Sachi Nandan Mohanty E.Laxmi Lydia Department of Management Information Systems Faculty of Economics and AdministrationKing Abdulaziz UniversityJeddahSaudi Arabia Department of business Administration Faculty of Economics and AdministrationKing Abdulaziz UniversityJeddah21589Saudi Arabia Department of Management Information Systems Faculty of Economics and AdministrationKing Abdulaziz UniversityJeddah21589Saudi Arabia School of Computer Science&Engineering(SCOPE) VIT-AP UniversityAmaravatiAndhra PradeshIndia Department of Computer Science and Engineering GMR Institute of TechnologyAndhra PradeshRajam 532127India

The current advancement in cloud computing,Artificial Intelligence(AI),and the Internet of Things(IoT)transformed the traditional healthcare system into smart *** services could be enhanced by incorporating key techniques like AI and *** convergence of AI and IoT provides distinct opportunities in the medical *** is regarded as a primary cause of death or post-traumatic complication for the ageing ***,earlier detection of older person falls in smart homes is required to improve the survival rate of an individual or provide the necessary ***,the emergence of IoT,AI,smartphones,wearables,and so on making it possible to design fall detection(FD)systems for smart home *** article introduces a new Teamwork Optimization with Deep Learning based Fall Detection for IoT Enabled Smart Healthcare systems(TWODLFDSHS).The TWODL-FDSHS technique’s goal is to detect fall events for a smart healthcare ***,the presented TWODL-FDSHS technique exploits IoT devices for the data collection ***,the TWODLFDSHS technique applies the TWO with Capsule Network(CapsNet)model for feature *** last,a deep random vector functional link network(DRVFLN)with an Adam optimizer is exploited for fall event detection.A wide range of simulations took place to exhibit the enhanced performance of the presentedTWODL-FDSHS *** experimental outcomes stated the enhancements of the TWODL-FDSHS method over other models with increased accuracy of 98.30%on the URFD dataset.

关键词： Internet of things smart healthcare deep learning team work optimizer capsnet fall detection

来源：评论

学校读者我要写书评

暂无评论

Improved Ingredients-based Recipe Recommendation Software using Machine Learning 11

Improved Ingredients-based Recipe Recommendation Software us...

引用

11th IEEE International Conference on Intelligent Computing and Information systems, ICICIS 2023

作者： Ashraf, Mariam Tarek, Khadija Fathy, Naglaa Mahmoud, Mariam Emad, Rana Khaled, Khadija Amr, Noran Ain Shams University Faculty of Computer and Information Science Software Engineering Department Cairo Egypt Ain Shams University Faculty of Computer and Information Science Information Systems Department Cairo Egypt

ISBN: (纸本)9798350322101

In this paper we propose an improved recipe recommendation system that employs image recognition of food ingredients. The system is currently a mobile application that performs image recognition on uploaded or camera-captured images and recommends recipes containing the recognized ingredients. We used the ResNet-V2 architecture to build a convolutional neural network model for image recognition, which was able to identify 33 different food ingredients with an accuracy rate of 89%. The recommendation system uses the identified ingredient labels, as well as user preferences and restrictions, to display a list of recipes containing the identified ingredients. This feature allows users to discover new and exciting recipes based on the ingredients they currently have at home, without having to worry about dietary restrictions or other preferences. Overall, our system provides a convenient and personalized way for users to discover and prepare delicious meals based on their unique needs and preferences. © 2023 IEEE.

关键词： Image recognition

来源：评论

学校读者我要写书评

暂无评论

A Plasmonic Wavelength Division Demultiplexer for 1310/1550/1650 nm Based on a Compact-Size Air-Gap Filter

A Plasmonic Wavelength Division Demultiplexer for 1310/1550/...

引用

2024 Laser science, LS 2024

作者： Wahsheh, Rami A. Wahsheh, Luay A. Communications Engineering Department Princess Sumaya University for Technology Amman Jordan Department of Computer Science and Information Systems University of North Georgia DahlonegaGA United States

We introduce a novel design and analysis of an ultra-compact plasmonic wavelength division demultiplexer using an air-gap filter. Our results indicate that the proposed design can sort 1310/1550 nm and 1310/1650 nm wa... 详细信息

关键词： Demultiplexing

来源：评论

学校读者我要写书评

暂无评论

CNN-LSTM: Development of Offline Signature Authentication 1

CNN-LSTM: Development of Offline Signature Authentication

引用

1st IEEE International Conference on Emerging Research in Computational science, ICERCS 2023

作者： Ganapathy, Revathy Lingeshwaran, D. Suthanthiradevi, P. Harish Kumar, M.V. Vels Insitute of Science Technology and Advanced Studies Department of Computer Science and Engineering Chennai India Srm Institute of Science and Technology Department of Data Science and Business Systems Chennai India

ISBN: (纸本)9798350359756

Offline Signature Authentication is a critical task in the field of document authentication, and its accuracy is essential for ensuring security while transactions. This research proposes two approaches: Initially Pre-trained CNN models are used to extract features from signature images, which are then combined with handcrafted features such as HOG and some other geometric features of signature. Such combined features are passed to bidirectional LSTM model in which drop out layer undergoes classification which differentiate real and forgery signature. The proposed system has potential applications in document authentication and security, subsequently combination of CNN models and additional features provides more comprehensive representation of signature images resulting in improved accuracy. Three signature datasets are utilized namely GDPS, CEDAR, and BHSig-Bengali each with varying signature styles and image quality. Our experimental outcomes reveal that Bidirectional Convolutional LSTM along with handcraft features attained maximum accuracy in offline signature verification system. © 2023 IEEE.

关键词： Long short-term memory

来源：评论

学校读者我要写书评

暂无评论

Deepfake Image Verification using DCNN with MobileNetV2 3

Deepfake Image Verification using DCNN with MobileNetV2

引用

3rd IEEE Delhi Section Flagship Conference, DELCON 2024

作者： Pujari, Jeevana Jyothi Bikku, Thulasi Prasanna, Kamepalli S. L. Kocherlakota, Sarvani Gupta, Nancy Anitha, Raju School of Computer Science and Engineering Vellore Institute of Technology Department of Database Systems Tamil Nadu Vellore India Amrita School of Computing Amaravati Amrita Vishwa Vidyapeetham Andhra Pradesh Amaravati India Amrita School of Business Amrita Vishwa Vidyapeetham Department of Business Andhra Pradesh Amaravati India School of Computer Science and Engineering Vellore Institute of Technology Department of Computer Science Core Tamil Nadu Vellore India Koneru Lakshmaiah Education Foundation Department of Computer Science and Engineering Guntur India

ISBN: (纸本)9798331518592

In this contemporary era we find so many things happening around us. They may or may not be rational, but we accept them and go on the path they show without knowing who is behind all that. This all is done by communication channels that make use of Artificial Intelligence to come up with content that looks as if some human being has done but it is a few lines of code that generate that content called Deepfake. This technology has been a boon for entertainment, fiction, art, etc. From our study, we found that there is research that found a deep fake image detection machine. This machine can adequately detect any changes made in the image. This has helped us go a bit closer to our goal by giving us creative ideas that ensure more authenticity and integrity of media. We can achieve our goal by using various fields of Deep Learning and computer Vision. DCNNs and MobileNetV2 have strong object detection identification and classification abilities and are perfect tools to help us get rid of the deepfake menace. Since technology is rapidly changing it brings new challenges with it. So, to cope with the ongoing challenges our system would have self-learning abilities and would effectively handle any deepfake methods and strategies. This system would serve as a mother to many other advanced deepfake detection systems that safeguard our visual media. Our target is to build a flexible, easy-to-use system that can effectively work in different environments and in collaboration with other software applications. These models would dominate deepfake expansion with 99% accuracy. Thus, our work would play a great role in protecting our data and help provide a safe and secure environment for every user. © 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

Prediction of room temperature in Trombe solar wall systems using machine learning algorithms

引用

Energy Storage and Saving 2024年第4期3卷 243-249页

作者： Seyed Hossein Hashemi Zahra Besharati Seyed Abdolrasoul Hashemi Seyed Ali Hashemi Aziz Babapoor Petroleum Systems Engineering Faculty of Engineering and Applied ScienceUniversity of ReginaReginaSK S4S 0A2Canada Department of Chemical Engineering University of Mohaghegh ArdabiliArdabil 13131-56199Iran Department of Chemistry School of ScienceAlzahra UniversityTehran ***Iran Computer Engineering Gachsaran Oil and Gas Exploitation CompanyGachsaran ***Iran Architectural Engineering Shiraz Islamic Azad UniversityShiraz ***Iran

A Trombe wall-heating system is used to absorb solar energy to heat *** parameters affect the system performance for optimal *** study evaluated the performance of four machine learning algorithms—linear regression,k-nearest neighbors,random forest,and decision tree—for predicting the room temperature in a Trombe wall *** accuracy of the algorithms was assessed using R^(2)and root mean squared error(RMSE)*** results demonstrated that the k-nearest neighbors and random forest algorithms exhibited superior performance,with R^(2)and RMSE values of 1 and *** contrast,linear regression and decision tree showed weaker *** findings highlight the potential of advanced machine learning algorithms for accurate room temperature prediction in Trombe wall systems,enabling informed design decisions to enhance energy efficiency.

关键词： Trombe wall Solar energy Thermal storage wall Machine learning algorithms

来源：评论

学校读者我要写书评

暂无评论

IoT-Aided Charity-Redistribution of Excess Food 2

IoT-Aided Charity-Redistribution of Excess Food

引用

2nd International Conference on Emerging Trends in Information Technology and engineering, ic-ETITE 2024

作者： Yogananth, V.S. Balasubramani, M. Vellore Institute of Technology Department of Computer Applications Tamil Nadu Vellore India School of Computer Science Engineering and Information Systems Vellore Institute of Technology Tamil Nadu Vellore India

ISBN: (纸本)9798350328202

The utilization of the Internet of Things (IOT) has shown significant potential in various aspects of daily life, yet its application in addressing social issues remains underdeveloped. India, with a substantial number of orphanages, faces challenges in efficiently distributing crucial information, hindering donors from obtaining accurate data on food reserves. This project proposes an IOT-based system to tackle this issue, using Node MCU IOT devices connected to sensors that relay data through a network, accessible and monitorable via web applications. The proposed system architecture aims to bridge the gap between those who waste food and those in poverty who desperately need it. We intend to facilitate food donations by implementing smart containers equipped with IOT technology. These containers can detect excess food waste and share this information via a GSM module with trusted charitable organizations. This application is designed to serve three main stakeholders: service providers, donors, and food recipients in need. By creating smart containers, we ensure that donated food reaches the right individuals. We prioritize collaborating with trusted charities to oversee the distribution process. Only authorized charity members can access and distribute the food stored in these containers, ensuring that it reaches those who need it most. © 2024 IEEE.

关键词： Containers

来源：评论

学校读者我要写书评

暂无评论

Automatic Speaker Recognition Using Mel-Frequency Cepstral Coefficients Through Machine Learning

引用

computers, Materials & Continua 2022年第6期71卷 5511-5521页

作者： U˘gur Ayvaz Hüseyin Gürüler Faheem Khan Naveed Ahmed Taegkeun Whangbo Abdusalomov Akmalbek Bobomirzaevich Department of Computer Engineering Istanbul Technical UniversityIstanbul34485Turkey Department of Information Systems Engineering Mugla Sitki Kocman UniversityMugla48000Turkey Artificial Intelligence Lab Department of Computer EngineeringGachon UniversitySeongnam13557Korea Department of Computer Science College of Computing and InformaticsUniversity of SharjahSharjah27272UAE

Automatic speaker recognition(ASR)systems are the field of Human-machine interaction and scientists have been using feature extraction and feature matching methods to analyze and synthesize these *** of the most commonly used methods for feature extraction is Mel Frequency Cepstral Coefficients(MFCCs).Recent researches show that MFCCs are successful in processing the voice signal with high *** represents a sequence of voice signal-specific *** experimental analysis is proposed to distinguish Turkish speakers by extracting the MFCCs from the speech *** the human perception of sound is not linear,after the filterbank step in theMFCC method,we converted the obtained log filterbanks into decibel(dB)features-based spectrograms without applying the Discrete Cosine Transform(DCT).A new dataset was created with converted spectrogram into a 2-D *** learning algorithms were implementedwith a 10-fold cross-validationmethod to detect the *** highest accuracy of 90.2%was achieved using Multi-layer Perceptron(MLP)with tanh activation *** most important output of this study is the inclusion of human voice as a new feature set.

关键词： Automatic speaker recognition human voice recognition spatial pattern recognition MFCCs spectrogram machine learning artificial intelligence

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：