检索结果-内蒙古大学图书馆

2024 IEEE International Conference on E-Health Networking, Application and Services, HealthCom 2024

作者： Karunarathna, Thilini Savindya Liang, Zilu Ubiquitous and Personal Computing Lab. Kyoto University of Advanced Science Kyoto Japan

ISBN: (纸本)9798350350548

Effective blood glucose regulation is essential for both diabetic and non-diabetic individuals. Finger-prick methods are inconvenient and painful, while continuous glucose monitoring (CGM) systems are costly and require frequent sensor replacements. This paper proposes an innovative approach for continuous real-time blood glucose level estimation using noninvasive physiological signals and ensemble regression techniques. We observed significant differences in the predictive power of individual features and overall model performance between the diabetic and the non-diabetic groups. The best model for the diabetic group achieved better performance (R2 = 0.50) compared to the best model for the non-diabetic group (R2 = 0.27). However, when evaluated for clinical accuracy, the non-diabetic models outperformed the diabetic models. The best non-diabetic model had 100% of predictions in the clinically acceptable zones of the Clarke Error Grids (CEG), while the best diabetic model had only 91%. The between-group differences in model performance are likely due to the variations in underlying glucose dynamics between diabetic and non-diabetic individuals. Our findings highlight the necessity of collecting and sharing large multimodal datasets from both diabetic and non-diabetic populations, as well as evaluating glucose prediction models from both mathematical and clinical accuracy. © 2024 IEEE.

关键词： Prediction models

来源：评论

学校读者我要写书评

暂无评论

An Efficacy Comparison of Supervised Machine Learning Classifiers for Cyberbullying Detection and Prediction

引用

International Journal of Bullying Prevention 2024年 1-20页

作者： Javed, Nashra Ahmed, Tasneem Faisal, Mohammad Advanced Computing & Research Lab. Department of Computer Application Integral University Uttar Pradesh Lucknow 226026 India

In the face of increasing cyberbullying incidents on online social networking sites (SNS), there is a growing demand for robust detection and prediction mechanisms to foster a safer digital environment. This paper investigates the performance of several supervised machine learning (ML) classifiers in terms of sentiment analysis and text categorization related to cyberbullying. It introduces frameworks for text pre-processing and classification with multiple cyberbullying classes like religion, age, gender, and ethnicity. Utilizing a comprehensive dataset of cyberbullying incidents collected from X (formerly Twitter), this study evaluates multiple supervised learning classifiers to identify the best one with the highest accuracy. The best-performing classifier is then used to develop a robust model for cyberbullying detection and prediction. In addition, feature extraction and model interpretability are achieved to learn more about how these classifiers make decisions when it comes to cyberbullying. The main aim of this research is to contribute practical solutions by identifying the most suitable machine learning models for diverse social networking sites datasets. Through performance evaluation, it is observed that the Random Forest classifier achieves an accuracy of 94%, outperforming the other classifiers and it is selected for the model development. Further, the developed model is trained, tested, and validated on different SNS platform datasets. Finally, the developed model is able to accurately perform detection and prediction of cyberbullying text as per the specified classes. This study has the potential to improve the model by applying deep learning and natural language processing (NLP) techniques to effectively detect cyberbullying in cyberspace. © The Author(s), under exclusive licence to Springer Nature Switzerland AG 2024.

关键词： Cyberbullying Deep learning Machine learning Natural language processing Sentiment analysis Social networking sites

来源：评论

学校读者我要写书评

暂无评论

LSTM-based Glucose Prediction for Prediabetes 13

LSTM-based Glucose Prediction for Prediabetes

引用

13th IEEE Global Conference on Consumer Electronic, GCCE 2024

作者： Carvalho, Cleber Franca Savindya Karunarathna, Thilini Liang, Zilu Kyoto University of Advanced Science Faculty of Engineering Kyoto Japan Kyoto University of Advanced Science Ubiquitous and Personal Computing Lab Kyoto Japan

ISBN: (纸本)9798350355079

Existing glucose prediction models were developed for diabetes populations. Their generalizability to other populations such as prediabetic population is unknown. To fill in the knowledge gap, we developed a glucose prediction model for prediabetic population. We evaluated the internal and external validity of the model. The results showed that the constructed model had good results for predicting glucose levels for prediabetic individuals, with mean absolute error (MAE), root mean square error (RMSE) and normalized RMSE (NRMSE) of 2.66, 4.00 and 0.21, respectively. When applied on other populations, the model achieved comparable performance on type 2 diabetes and better performance on type 1 diabetes in terms of NRMSE. © 2024 IEEE.

关键词： Prediction models

来源：评论

学校读者我要写书评

暂无评论

Deep Learning Compiler Optimization on Multi-Chiplet Architecture 5

Deep Learning Compiler Optimization on Multi-Chiplet Archite...

引用

5th IEEE International Conference on Artificial Intelligence Circuits and Systems, AICAS 2023

作者： Xu, Huiqing Mao, Kuang Pan, Quihong Tang, Zhaorong Wang, Mengdi Wang, Ying Zhejiang Lab Research Center for Advanced Computing Systems Hangzhou China Chinese Academy of Sciences Institute of Computing Technology Beijing China

ISBN: (纸本)9798350332674

Multi-chiplet architecture can provide a high-performance solution for new tasks such as deep learning models. In order to fully utilize chiplets and accelerate the execution of deep learning models, we present a deep learning compilation optimization framework for chiplets, and propose a scheduling method based on data dependence. Experiments show that our method improves the compilation efficiency, and the performance of the scheduling scheme is at least 1-2 times higher than the traditional algorithms. © 2023 IEEE.

关键词： Deep learning

来源：评论

学校读者我要写书评

暂无评论

FedCars: An Efficient Scheduling Framework for In-Vehicle Federated Learning Integrated With CARLA-Cosimulation

引用

IEEE Transactions on Vehicular Technology 2024年第5期74卷 7997-8011页

作者： Hong, Yuncong Lv, Bojie Wang, Shuai Wang, Rui Tan, Haisheng Lau, Francis C.M. China Hong Kong Department of Electrical and Electronic Engineering Shenzhen China Shenzhen Institute of Advanced Technology Shenzhen518055 China LINKE Lab Hefei China Department of Computing Science Hong Kong

Federated learning on vehicles has been widely considered in autonomous driving. The model uploading of federated learning is time-consuming, which motivates the transmission optimization that exploits the mobility of vehicles. In this paper, an optimization framework integrated with the high-fidelity traffic simulator CARLA, namely FedCars, is proposed for the scheduling of model uploading in a Vehicle-to-Infrastructure (V2I) network. Specifically, a group of data collection vehicles (DCVs) move along their planned routes with random velocities to collect sensor data and train a model via federated learning, where the uplink model transmission of multiple DCVs in one learning iteration is jointly optimized. We first develop the FedCars simulator from CARLA, to obtain a high-fidelity trajectory dataset of DCVs in the considered traffic scenario, and train the transition probabilities of the DCVs' locations. Hence, the uplink transmission time and power of all DCVs in all the time slots can be formulated as a finite-horizon Markov Decision Process (MDP), where the prediction of DCVs' future locations is exploited to minimize a weighted sum of the average model uploading time and the average energy consumption. A low-complexity FedCars optimizer with a non-trivial performance bound is then proposed, where the optimization consists of global transmission time allocation and local power allocation. The former periodically allocates the transmission time for the remaining time slots according to an average trajectory. Hence, the power allocation of all DCVs can be decoupled to local finite-horizon MDPs in the latter. They can be solved by one-step policy improvement with low computational complexity. The simulation results show that the proposed solution framework can achieve both a significant performance gain over the baselines and a flexible trade-off between performance and complexity. © 1967-2012 IEEE.

关键词： Markov processes

来源：评论

学校读者我要写书评

暂无评论

User Authentication with Keystroke Dynamics: Performance Evaluation in Neural Network 6

User Authentication with Keystroke Dynamics: Performance Eva...

引用

6th IEEE International Conference on Artificial Intelligence in Engineering and Technology, IICAIET 2024

作者： Kek, Xin-Jin Leau, Yu-Beng Tan, Soo Fun Cybersecurity Research Lab Faculty of Computing and Informatics Universiti Malaysia Sabah Sabah Malaysia Creative Advanced Machine Intelligence Research Centre Faculty of Computing and Informatics Universiti Malaysia Sabah Sabah Malaysia

ISBN: (纸本)9798350389692

In the current context of research, it investigates user authentication through keystroke dynamics using the neural network model which focuses on Convolutional Neural Network (CNN), Gated Recurrent Unit (GRU) and Long Short-Term (LSTM). As of above, four distinct experiments were conducted: (1) evaluating 3 different algorithms, (2) layering configuration for CNN and GRU fusion, (3) number of filters and neurons as hyperparameters, (4) assessing combinations of events counts and classes. In justification, the final outcome demonstrates an achieved accuracy of 87%. Based on the experiments that were carried out, optimal hyperparameters were identified: window size of 40, 256 for the number of filters and neurons and 2500 as the optimal number of events with 30 as the number of classes. These findings underscore a significance of considering diverse combinations of a number of events and classes by providing insights into the determination of optimal hyperparameters for keystroke dynamics in performance evaluation. © 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

High-throughput Sampling, Communicating and Training for Reinforcement Learning Systems 31

High-throughput Sampling, Communicating and Training for Rei...

引用

31st IEEE/ACM International Symposium on Quality of Service, IWQoS 2023

作者： Zhao, Laiping Dai, Xinan Zhao, Zhixin Xin, Yusong Hu, Yitao Qian, Jun Yao, Jun Li, Keqiu College of Intelligence & Computing Tianjin University Tianjin Key Lab. of Advanced Networking Tianjin China Huawei Noahs Ark Lab Beijing China

ISBN: (纸本)9798350399738

Reinforcement Learning (RL) algorithms require large amounts of computational resources and time to train due to the simultaneous model training and real-time interactions with simulation environments. This results in an RL algorithm's training time potentially exceeding days to months. We present HRL, a comprehensive optimization system designed to improve the system throughput of RL algorithms. HRL addresses the challenges of discovering and identifying bottlenecks in the sampling, training, or communication stages and promptly improving their performance. We propose a group-parallel pipeline method to improve the sampling efficiency and apply data quantization and multi-learner training to resolve the network and learning bottlenecks. The HRL system has been fully implemented and integrated with XingTian. The results of the experiments show that the HRL system can improve the throughput by 18.6%-90.6%. © 2023 IEEE.

关键词： Pipelines

来源：评论

学校读者我要写书评

暂无评论

Enhancing Electronic Document Security with Lightweight Digital Signature 6

Enhancing Electronic Document Security with Lightweight Digi...

引用

6th IEEE International Conference on Artificial Intelligence in Engineering and Technology, IICAIET 2024

作者： Fun, Tan Soo Ahmad Zulkifli, Nur Sufia Binti Sia, Florence Hung, Lai Po Universiti Malaysia Sabah Cybersecurity Research Lab Faculty of Computing and Informatics Sabah Kota Kinabalu Malaysia Universiti Malaysia Sabah Creative Advanced Machine Intelligence Research Centre Faculty of Computing and Informatics Sabah Kota Kinabalu Malaysia

ISBN: (纸本)9798350389692

The heightened frequency and ferocity of recent security breaches and electronic document falsification pose significant concerns globally. These incidents underscore the critical need for robust security measures to protect sensitive documents from unauthorized access and tampering. While digital signatures are essential for ensuring the authenticity and integrity of electronic documents, traditional digital signature algorithms can be computationally intensive, making them unsuitable for resource-constrained environments such as IoT devices and mobile applications. This paper explores lightweight digital signature algorithms that offer robust security with minimal resource consumption. We investigate, implement, and evaluate the performance evaluation of lightweight digital signatures, highlighting their efficiency in both signature generation and verification processes. Our findings demonstrate that these algorithms can significantly enhance the security of electronic documents without compromising performance. © 2024 IEEE.

关键词： Electronic document identification systems

来源：评论

学校读者我要写书评

暂无评论

Brain Network Features Differentiate Intentions from Different Emotional Expressions of the Same Text 48

Brain Network Features Differentiate Intentions from Differe...

引用

48th IEEE International Conference on Acoustics, Speech and Signal Processing, ICASSP 2023

作者： Li, Zhongjie Zhao, Bin Zhang, Gaoyan Dang, Jianwu Tianjin University Cca Lab College of Intelligence and Computing Tianjin China Japan Advanced Institute of Science and Technology Information Science Japan

ISBN: (纸本)9781728163277

Intent differentiation in speech communication relies not only on linguistic information but also on paralinguistic information. The same textual content, when pronounced with different prosodies and emotions, may express totally different intentions. The true intentions in this condition can be easily grasped by our brain. Therefore, combining text, speech, and electroencephalography (EEG) for intent discrimination on the same text may be an effective approach. Before fusing speech and text modalities, the current study focused on exploring effective EEG-based features for Chinese intent recognition as no previous research has utilized EEG signals for this purpose. To tackle this issue, we first created a Chinese multimodal spoken language intention understanding (CMSLIU) dataset, in which the same texts were pronounced with varying prosodies to express different intents. To identify effective brain features that were most relevant to intent recognition improvement, we compared the event-related spectral perturbation and effective brain connectivity patterns on two intent conditions (praise vs. irony). It was found that the praise expression tended to elicit stronger high-frequency brain activities while the irony expression involved a more suppressive network connection in the right hemisphere. These features were trained on the CMSLIU dataset and achieved an intention classification accuracy of 78.66%, which indicated a great potential of the EEG features in intent discrimination on the same text. © 2023 IEEE.

关键词： brain network features EEG Human-computer interaction Intent understanding paralinguistic information

来源：评论

学校读者我要写书评

暂无评论

Stereoselective recognition of morphine enantiomers by μ-opioid receptor

引用

National Science Review 2024年第3期11卷 183-191页

作者： Yibo Wang Van A.Ngo Xiaohui Wang Laboratory of Chemical Biology Changchun Institute of Applied ChemistryChinese Academy of Sciences Advanced Computing for Life Sciences and Engineering Group Science Engagement Section National Center for Computational Sciences Oak Ridge National Lab School of Applied Chemistry and Engineering University of Science and Technology of China Beijing National Laboratory for Molecular Sciences

Stereospecific recognition of chiral molecules plays a crucial role in biological systems. The μ-opioid receptor（MOR） exhibits binding affinity towards（-）-morphine, a well-established gold standard in pain management, while it shows minimal binding affinity for the（+）-morphine enantiomer, resulting in a lack of analgesic activity. Understanding how MOR stereoselectively recognizes morphine enantiomers has remained a puzzle in neuroscience and pharmacology for over half-a-century due to the lack of direct observation techniques. To unravel this mystery, we constructed the binding and unbinding processes of morphine enantiomers with MOR via molecular dynamics simulations to investigate the thermodynamics and kinetics governing MOR's stereoselective recognition of morphine enantiomers. Our findings reveal that the binding of（-）-morphine stabilizes MOR in its activated state, exhibiting a deep energy well and a prolonged residence time. In contrast,（+）-morphine fails to sustain the activation state of ***, the results suggest that specific residues, namely D1142.50and D1473.32, are deprotonated in the active state of MOR bound to（-）-morphine. This work highlights that the selectivity in molecular recognition goes beyond binding affinities, extending into the realm of residence time.

关键词： stereoselective recognition thermodynamics and kinetics residence time morphine enantiomers μ-opioid receptor

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：