检索结果-内蒙古大学图书馆

作者： Shen, Xiangyu Jiang, Hongbo Liu, Daibo Yang, Kehua Deng, Feiyang Lui, John C. S. Liu, Jiangchuan Dustdar, Schahram Luo, Jun Hunan University College of Computer Science and Electronic Engineering Hunan Changsha410082 China The City University of Hong Kong School of Graduate Studies Hong Kong The Chinese University of Hong Kong Computer Science and Engineering Department Hong Kong Simon Fraser University School of Computing Science BurnabyBCV5A 1S6 Canada Vienna University of Technology Distributed Systems Group Vienna1040 Austria Nanyang Technological University School of Computer Engineering 639798 Singapore

As mobile shopping has gradually become the mainstream shopping mode, recommendation systems are gaining an increasingly wide adoption. Existing recommendation systems are mainly based on explicit and implicit user behaviors. However, these user behaviors may not directly indicate users' inner feelings, causing erroneous user preference estimation and thus leading to inaccurate recommendations. Inspired by our key observation on the correlation between pupil size and users' inner feelings, we consider using the change of pupil size when browsing to model users' preferences, so as to achieve targeted recommendations. To this end, we propose PupilRec as a computer-vision-based recommendation framework involving a mobile terminal and a server side. On the mobile terminal, PupilRec collects users' pupil size change information through the front camera of smartphones;it then preprocesses the raw pupil size data before transmitting them to the server. On the server side, PupilRec utilizes the Tsfresh package and Random Forest algorithm to figure out the key time-series features directly implying user preferences. PupilRec then trains a neural network to fit a user preference model. Using this model, PupilRec predicts user preference to obtain a user-product matrix and further simplifies it by singular value decomposition. Finally, the real-time recommendation is achieved by a collaborative filtering module that retrieves recommended contents to users smartphones. We prototype PupilRec and conduct both experiments and field studies to comprehensively evaluate the effectiveness of PupilRec by recruiting 67 volunteers. The overall results show that PupilRec can accurately estimate users' preference and can recommend products users interested in. © 2014 IEEE.

关键词： Smartphones

来源：评论

学校读者我要写书评

暂无评论

Hardware Realization of High-Speed Area-Efficient Floating Point Arithmetic Unit on FPGA

Hardware Realization of High-Speed Area-Efficient Floating P...

引用

Machine Intelligence and Smart Innovation (ICMISI), International Conference on

作者： Mohammed H. Yacoub Samar M. Ismail Lobna A. Said Nanoelectronics Integrated Systems Center (NISC) Nile University Giza Egypt School of Engineering and Computer Science University of Hertfordshire (UH-GAF) Egypt

ISBN: (数字)9798350365740

ISBN: (纸本)9798350365757

Floating point representations are required in many applications due to their universality and ability to represent huge numbers accurately and in compact bit-width. Floating point arithmetic is complex, performance inefficient, and area-consuming compared to integer arithmetic operations. In this paper, hardware realization of area-efficient high-performance floating point arithmetic units for IEEE 754 floating point single precision and double precision formats on FPGA are proposed. The proposed units achieved the same accuracy as software in all tested cases and were able to produce the same chaotic behavior of the Ro¨ssler system identical to the software simulation results on MATLAB. The hardware realization of the floating point adder/subtractor occupied only 0.11% and 0.28% of AMD's KCU105 FPGA while achieving a maximum frequency of 121 MHz and 106 MHz for the single and double precision units respectively. The floating point multipliers on the other hand occupied 0.14% and 0.04% of the total area and reached a maximum frequency of 140.5 MHz and 104.8 MHz for the single and double precision respectively.

关键词： Technological innovation Simulation Time series analysis Hardware Software Floating-point arithmetic Field programmable gate arrays

来源：评论

学校读者我要写书评

暂无评论

Evaluation of Predictive Display for Teleoperated Driving Using CARLA Simulator

Evaluation of Predictive Display for Teleoperated Driving Us...

引用

IEEE/RSJ International Conference on Intelligent Robots and systems (IROS)

作者： Fatima Kashwani Bilal Hassan Peng-Yong Kong Majid Khonji Jorge Dias Center for Autonomous Robotic Systems (KUCARS) Abu Dhabi UAE Department of Electrical Engineering and Computer Science Khalifa University Abu Dhabi UAE

ISBN: (数字)9798350377705

ISBN: (纸本)9798350377712

Before the world-wide deployment of autonomous vehicles, it is essential to implement intermediate solutions with partial autonomy. One such solution is the use of vehicle teleoperation, the act of controlling a vehicle from a distance. In real time applications of teleoperation, it is often pertinent to use augmented reality components within the teleoperator view, which are referred to as a predictive display. In this work, we evaluate our predictive display method, which is a guiding path based on the free space in the environment. The path is generated based on our Dual Transformer Network (DTNet), which uses both object detection and lane semantic segmentation to define the free space in the environment. While the model has previously performed well on image data, it is necessary to observe its accuracy in the presence of time delay and packet loss, to assess its performance in a real-time setting. Thus, in this work, we use CARLA simulator to compare the detected free space on the teleoperator side to the true free space on the vehicle side across different values of time delay and packet loss. Under optimal network conditions, our model yielded a remarkable 87.9% DSC score and 81.3% IoU score. Defining our minimum performance threshold as 80% DSC and 70% IoU, we conclude that our model can effectively mitigate the challenges of time delay below 100ms and packet loss below 1%, both of which represent substantial tolerances.

关键词： Space vehicles Teleoperators Visualization Accuracy Delay effects Packet loss Transformers Real-time systems Vehicle dynamics Usability

来源：评论

学校读者我要写书评

暂无评论

IOT based wearable sensor system architecture for classifying human activity

引用

Measurement: Sensors 2025年 39卷

作者： Mahalakshmi, V. Kumar, Pramod Bhende, Manisha Keshta, Ismail Rathod, Swatiben Yashvantbhai Ramesh, Janjhyam Venkata Naga Department of Computer Science College of Engineering and computer science Jazan University Jazan45142 Saudi Arabia Computer science and Technology Ganga Institute of Technology and Management Kablana Haryana Jhajjar124104 India Dr. D. Y. Patil Vidyapeeth Pune India Dr. D. Y. Patil School of Science & Technology Tathawade Pune India Computer Science and Information Systems Department College of Applied Sciences AlMaarefa University Riyadh Saudi Arabia Instrumentation And Control Department Vishwakrma Government Engineer College Chandkheda Ahmedabad382424 India Department of Computer Science and Engineering Koneru Lakshmaiah Education Foundation Vaddeswaram Guntur Dist. Andhra Pradesh 522302 India

Human Activity Recognition (HAR) has applications in diverse fields, including sports management and behavior classification. Existing HAR methods can be categorized into three main approaches: camera-based, wearable sensor-based, and Wi-Fi sensing-based. Camera-based methods suffer from privacy concerns, while wearable sensor-based methods face limitations in battery longevity and continuous monitoring. Wi-Fi sensing methods mitigate privacy and battery issues but rely on costly Intel 5300 network cards or software-defined radio (SDR) platforms, limiting scalability. This paper presents a cost-effective IoT-based human activity recognition system using ESP32, leveraging its Wi-Fi sensing capabilities. The proposed system follows a structured workflow: (i) channel state information (CSI) is extracted from ESP32 modules, (ii) data preprocessing is performed using Hampel and Gaussian filters for noise and outlier removal, (iii) dimensionality reduction is achieved through Principal Component Analysis (PCA) and Discrete Wavelet Transform (DWT), and (iv) activity classification is conducted using Dynamic Time Warping (DTW) and the K-Nearest Neighbors (KNN) algorithm. Experimental evaluations demonstrate that the proposed system achieves an average recognition accuracy of 98.6 % across six human activities, comparable to high-end Intel 5300-based HAR systems, while significantly reducing hardware costs and improving ease of deployment. © 2025

关键词： Discrete wavelet transforms

来源：评论

学校读者我要写书评

暂无评论

The Effectiveness of Local Updates for Decentralized Learning under Data Heterogeneity

arXiv

引用

arXiv 2024年

作者： Wu, Tongle Li, Zhize Sun, Ying School of Electrical Engineering and Computer Science Pennsylvania State University United States School of Computing and Information Systems Singapore Management University Singapore

We revisit two fundamental decentralized optimization methods, Decentralized Gradient Tracking (DGT) and Decentralized Gradient Descent (DGD), with multiple local updates. We consider two settings and demonstrate that incorporating local update steps can reduce communication complexity. Specifically, for µ-strongly convex and L-smooth loss functions, we proved that local DGT achieves communication complexity O∼(µ(KL+1) + µ(1δ+−µρ) + (1−ρρ)2 · Lµ+δ ) , where K is the number of additional local update, ρ measures the network connectivity and δ measures the second-order heterogeneity of the local losses. Our results reveal the tradeoff between communication and computation and show increasing K can effectively reduce communication costs when the data heterogeneity is low and the network is well-connected. We then consider the over-parameterization regime where the local losses share the same minimums. We proved that employing local updates in DGD, even without gradient correction, achieves exact linear convergence under the Polyak-Lojasiewicz (PL) condition, which can yield a similar effect as DGT in reducing communication complexity. Customization of the result to linear models is further provided, with improved rate expression. Numerical experiments validate our theoretical results. Copyright © 2024, The Authors. All rights reserved.

关键词： Decentralized control

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

Improved Ingredients-based Recipe Recommendation Software us...

引用

IEEE International Conference on Intelligent Computing and Information systems (ICICIS)

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

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.

关键词：

来源：评论

学校读者我要写书评

暂无评论

Role of RPA & AI in Optimizing Network Field Services 7

Role of RPA & AI in Optimizing Network Field Services

引用

7th IEEE International Conference on Computational systems and Information Technology for Sustainable Solutions, CSITSS 2023

作者： Yendluri, Dileep Kumar Ponnala, Jainath Tatikonda, Reshmi Kempanna, M. Thatikonda, Ramya Bhuvanesh, A. University of the Cumberlands Computer Science United States Jawaharlal Technological University Ece Hyderbad India University of the Central Missouri Computer Information Systems Warrensburg United States Bangalore Institute of Technology Department of Artificial Intelligence and Machine Learning Karnataka Bengaluru India University of the Cumberlands Information Technology Williamsburg United States Psn College of Engineering and Technology Department of Electrical and Electronics Engineering Tamil Nadu Tirunelveli India

ISBN: (纸本)9798350343144

The paper investigates incorporating and implementing RPA and AI technologies within NFS to improve efficiency and boost service quality. Robotic Process Automation enables the streamlining of repetitive processes. It simplifies work processes and releases personnel for more critical projects. On the contrary, AI strengthens NFS providers via data-informed decision-making. Additionally, it facilitates anticipatory maintenance and preemptive network administration. The document explores particular RPA and cognitive computing instances, including AI-powered network issue resolution and AI-driven preventive maintenance for the network hardware. Furthermore, it emphasizes the advantages and constraints of adopting robotic process automation and artificial intelligence within the NFS framework. It considers aspects including data handling, growth potential, and moral implications. Prospects for the future related to RPA and AI within the NFS domain are also analyzed. It foresees self-governing network management, sophisticated predictive analytics, and merging with IoT and edge computing technologies. The article highlights the revolutionary capability of robotic process automation and artificial intelligence in transforming the field service sector, aiming for increased efficiency, reliability, and a customer-centric approach. It emphasizes embracing these technologies to keep up in the dynamic business sector. © 2023 IEEE.

关键词： Decision making

来源：评论

学校读者我要写书评

暂无评论

Multi-stage biomedical feature selection extraction algorithm for cancer detection

引用

SN Applied sciences 2023年第5期5卷 131页

作者： Keshta, Ismail Deshpande, Pallavi Sagar Shabaz, Mohammad Soni, Mukesh Bhadla, Mohit kumar Muhammed, Yasser Computer Science and Information Systems Department College of Applied Sciences AlMaarefa University Riyadh Saudi Arabia College of Engineering Pune India Arba Minch University Arba Minch Ethiopia Department of CSE University Centre for Research & Development Chandigarh University Punjab Mohali140413 India Department of Information Technology Ahmedabad Institute of Technology Ahmedabad India College of Technical Engineering Al-Farahidi University Baghdad Iraq

Cancer is a significant cause of death worldwide. Early cancer detection is greatly aided by machine learning and artificial intelligence (AI) to gene microarray data sets (microarray data). Despite this, there is a significant discrepancy between the number of gene features in the microarray data set and the number of samples. Because of this, it is crucial to identify markers for gene array data. Existing feature selection algorithms, however, generally use long-standing, are limited to single-condition feature selection and rarely take feature extraction into account. This work proposes a Multi-stage algorithm for Biomedical Deep Feature Selection (MBDFS) to address this issue. In the first, three feature selection techniques are combined for thorough feature selection, and feature subsets are obtained;in the second, an unsupervised neural network is used to create the best representation of the feature subset to enhance final classification accuracy. Using a variety of metrics, including a comparison of classification results before and after feature selection and the performance of alternative feature selection methods, we evaluate MBDFS's efficacy. The experiments demonstrate that although MBDFS uses fewer features, classification accuracy is either unchanged or enhanced. © 2023, The Author(s).

关键词： Feature Selection

来源：评论

学校读者我要写书评

暂无评论

Q-learning-based Joint Design of Adaptive Modulation and Precoding for Physical Layer Security in Visible Light Communications

Q-learning-based Joint Design of Adaptive Modulation and Pre...

引用

IEEE Conference on Vehicular Technology (VTC)

作者： Duc M. T. Hoang Thanh V. Pham Anh T. Pham Chuyen T. Nguyen Hanoi University of Science and Technology Vietnam Department of Mathematical and Systems Engineering Shizuoka University Japan Department of Computer Science and Engineering The University of Aizu Japan

Physical layer security (PLS) offers a unique approach to protecting information confidentiality against eavesdropping by malicious users. This paper studies a joint design of adaptive M−ary pulse amplitude modulation (PAM) and precoding for performance improvement of PLS in visible light communications (VLC). It is known that higher-order modulation results in a better secrecy capacity at the expense of a higher bit-error rate (BER). On the other hand, a proper precoding design can also enhance secrecy performance. The proposed design, therefore, aims at the optimal PAM modulation order and precoder to maximize a utility function that takes into account the secrecy capacity and BERs of the legitimate user (Bob)’s and the eavesdropper (Eve)’s channel. Due to the lack of a closed-form expression for the utility function, a Q-learning-based design is proposed and evaluated. Compared to the non-adaptive approach under all different settings of Bob’s and Eve’s positions, simulation results verify that the proposed joint adaptive design achieves a good balance between the secrecy capacity and BER of Bob’s channel while maintaining a sufficiently high BER of Eve’s channel.

关键词：

来源：评论

学校读者我要写书评

暂无评论

Software Defect Prediction Using Ensemble Technique

Software Defect Prediction Using Ensemble Technique

引用

International Conference on Information Management and Technology (ICIMTech)

作者： Pandu Wicaksono Panji Arisaputra Rezky Yunanda Willy Kristian Almuzhidul Mujhid Software Engineering Department School of Computer Science Bina Nusantara University Jakarta Indonesia Information Systems Department School of Information Systems Bina Nusantara University Jakarta Indonesia

ISBN: (数字)9798350390025

ISBN: (纸本)9798350390032

Software defect prediction is the methodical process of identifying code segments that are likely to have problems. This is done by analyzing software metrics and using categorization algorithms. This work introduces an alternate approach that utilizes ensemble learning techniques to improve the effectiveness of fault detection models. The models utilized in this study are CodeGPT and CodeBERT, both of which are transformer-based deep learning models that can extract code features from source code. The models are trained using the PROMISE dataset, which consists of Java projects that have been labeled with defects. The test results show that using ensemble learning methods can improve the accuracy, precision, recall, and F1 score of a single model by around 1-3%. This study contributes to the progress of software dependability and quality by utilizing advanced software fault prediction algorithms.

关键词： Codes Source coding Software algorithms Predictive models Prediction algorithms Feature extraction Transformers Software Ensemble learning Software measurement

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：