检索结果-内蒙古大学图书馆

A new online field feature selection algorithm based on streaming data

Journal of Ambient Intelligence and Humanized Computing 2024年第2期15卷 1365-1377页

作者： Zhang, Zhenjiang Song, Fuxing Zhang, Peng Chao, Han-Chieh Zhao, Yingsi School of Software Engineering Beijing Jiaotong University Beijing100044 China Department of Electronic and Information Engineering Key Laboratory of Communication and Information Systems Beijing Municipal Commission of Education Beijing Jiaotong University Beijing100044 China School of Information Science and Engineering Fujian University of Technology Fuzhou350118 China School of Mathematics and Computer Science Wuhan Polytechnic University Wuhan430023 China Department of Electrical Engineering National Dong Hwa University Hualien97401 Taiwan School of Economics and Management Beijing Jiaotong University Beijing100044 China

The rapid development of Internet technology derived out a massive network text data. Therefore, how to classify the massive text data efficiently has important theoretical significance and application value. In order to acquire accurate classification results, the process has been divided into two parts. In terms of text representation, this paper proposes an online field feature selection algorithm (OFFS algorithm) based on streaming data, which solves the problems of low efficiency and memory consumption of traditional feature selection algorithms. With improvements in the vector space model, the new algorithm can select the real-time feature of the data and quickly generate text vector. In the aspect of classifier design, an OFFS-BP neural network text classifier based on BP neutral network and OFFS algorithm is designed. It adapts to the distributed parallel computing, reduces the training time and balances the computation efficiency and classification accuracy. Finally based on the Spark platform, the OFFS-BP neural network classifier is implemented. The experimental results show that the OFFS-BP neural network classifier is more suitable for big data environment with less computation time and higher classification efficiency. © Springer-Verlag GmbH Germany, part of Springer Nature 2018.

关键词： Neural networks

来源：评论

学校读者我要写书评

暂无评论

PD-Refiner: An Underlying Surface Inheritance Refiner with Adaptive Edge-Aware Supervision for Point Cloud Denoising 24

PD-Refiner: An Underlying Surface Inheritance Refiner with A...

引用

32nd ACM International Conference on Multimedia, MM 2024

作者： Zhang, Chengwei Zhang, Xueyi Yue, Xianghu Lao, Mingrui Jiang, Tao Wang, Jiawei Zhang, Fubo Chen, Longyong National Key Laboratory of Microwave Imaging Aerospace Information Research Institute Chinese Academy of Sciences Beijing China Laboratory for Big Data and Decision National University of Defense Technology Hunan Changsha China Department of Electrical and Computer and Engineering National University of Singapore Singapore Singapore National Key Laboratory of Information Systems Engineering National University of Defense Technology Hunan Changsha China School of Software Engineering Tongji University Shanghai China

ISBN: (纸本)9798400706868

Point clouds from real-world scenarios inevitably contain complex noise, significantly impairing the accuracy of downstream tasks. To tackle this challenge, cascading encoder-decoder architecture has become a conventional technical route to iterative denoise. However, circularly feeding the output of denoiser as its input again involves the re-extraction of underlying surface, leading to unstable denoising process and over-smoothed geometric details. To address these issues, we propose a novel denoising paradigm dubbed PD-Refiner that employs a single encoder to model the underlying surface. Then, we leverage several lightweight hierarchical Underlying Surface Inheritance Refiners (USIRs) to inherit and strengthen it, thereby avoiding the re-extraction from the intermediate point cloud. Furthermore, we design adaptive edge-aware supervision to improve the edge awareness of the USIRs, allowing for the adjustment of the denoising preferences from global structure to local details. The results demonstrate that our method not only achieves state-of-the-art performance in terms of denoising stability and efficacy, but also enhances edge clarity and point cloud uniformity. © 2024 ACM.

关键词： edge-aware supervision

来源：评论

学校读者我要写书评

暂无评论

Structured Dialogue System for Mental Health: An LLM Chatbot Leveraging the PM+ Guidelines 16th

Structured Dialogue System for Mental Health: An LLM Chatbo...

引用

16th International Conference on Social Robotics, ICSR + InnoBiz 2024

作者： Chen, Yixiang Zhang, Xinyu Wang, Jinran Xie, Xurong Yan, Nan Chen, Hui Wang, Lan Guangdong-Hong Kong-Macao Joint Laboratory of Human-Machine Intelligence-Synergy Systems Shenzhen Institute of Advanced Technology Chinese Academy of Sciences Shenzhen China University of Chinese Academy of Sciences Beijing China Institute of Software Chinese Academy of Sciences Beijing China East China Normal University Shanghai China Wuhan Research Institute of Posts and Telecommunications Wuhan China

ISBN: (纸本)9789819611508

The Structured Dialogue System, referred to as SuDoSys, is an innovative Large Language Model (LLM)-based chatbot designed to provide psychological counseling. SuDoSys leverages the World Health Organization (WHO)’s Problem Management Plus (PM+) guidelines to deliver stage-aware multi-turn dialogues. Existing methods for employing an LLM in multi-turn psychological counseling typically involve direct fine-tuning using generated dialogues, often neglecting the dynamic stage shifts of counseling sessions. Unlike previous approaches, SuDoSys considers the different stages of counseling and stores essential information throughout the counseling process, ensuring coherent and directed conversations. The system employs an LLM, a stage-aware instruction generator, a response unpacker, a topic database, and a stage controller to maintain dialogue flow. In addition, we propose a novel technique that simulates counseling clients to interact with the evaluated system and evaluate its performance automatically. When assessed using both objective and subjective evaluations, SuDoSys demonstrates its effectiveness in generating logically coherent responses. The system’s code and program scripts for evaluation are open-sourced (https://***/EthanLifeGreat/SuDoSys). © The Author(s), under exclusive license to Springer Nature Singapore Pte Ltd. 2025.

关键词： Modeling languages

来源：评论

学校读者我要写书评

暂无评论

Architecting Digital Twins for Intelligent Transportation systems

arXiv

引用

arXiv 2025年

作者： Bhatt, Hiya Sahil Vaidhyanathan, Karthik Biju, Rahul Gangadharan, Deepak Trestian, Ramona Shah, Purav Software Engineering Research Centre IIIT Hyderabad India Computer Systems Group IIIT Hyderabad India Faculty of Science and Technology Middlesex University London United Kingdom

Modern transportation systems face growing challenges in managing traffic flow, ensuring safety, and maintaining operational efficiency amid dynamic traffic patterns. Addressing these challenges requires intelligent solutions capable of real-time monitoring, predictive analytics, and adaptive control. This paper proposes an architecture for DigIT, a Digital Twin (DT) platform for Intelligent Transportation systems (ITS), designed to overcome the limitations of existing frameworks by offering a modular and scalable solution for traffic management. Built on a Domain Concept Model (DCM), the architecture systematically models key ITS components enabling seamless integration of predictive modeling and simulations. The architecture leverages machine learning models to forecast traffic patterns based on historical and real-time data. To adapt to evolving traffic patterns, the architecture incorporates adaptive Machine Learning Operations (MLOps), automating the deployment and lifecycle management of predictive models. Evaluation results highlight the effectiveness of the architecture in delivering accurate predictions and computational efficiency. © 2025, CC BY.

关键词： Adaptive control systems

来源：评论

学校读者我要写书评

暂无评论

An Empirical Study on Edge-to-Cloud Continuum for Smart Applications: Performance, Design Patterns, and Key Factors

An Empirical Study on Edge-to-Cloud Continuum for Smart Appl...

引用

IEEE International Conference on Edge Computing (EDGE)

作者： Norman Chen Adel N. Toosi Bahman Javadi Daghash Alqahtani Mohammad S. Aslanpour Minxian Xu Department of Software Systems and Cybersecurity Monash University Clayton Australia School of Computer Data and Mathematical Sciences Western Sydney University Chinese Academy of Sciences Shenzhen Institute of Advanced Technology

ISBN: (数字)9798350368499

ISBN: (纸本)9798350368505

The rapid evolution of cloud-native technologies has facilitated seamless application deployment and execution across the entire edge-to-cloud continuum. This continuum offers a myriad of benefits, including reduced latency, optimized bandwidth utilization, enhanced data privacy, improved reliability, scalability, and flexibility. However, realizing a coherent edge-to-cloud continuum poses challenges especially in resource management, due to the heterogeneous and dynamic nature of computing resources such as resource scheduling and load balancing. This paper focuses on the Container-as-a-Service model enabling independent execution of functions/microservices anywhere on the continuum. We propose an architectural design for constructing a practical edge-to-cloud infrastructure and conduct comprehensive performance evaluations using a real edge-to-cloud testbed. Through an empirical study, we aim to identify key factors impacting application performance and resource management within the continuum, with a specific focus on AI-based IoT applications. Our experiments explore various design patterns including load balancing techniques, scheduling algorithms, invocation methods, gateway and data source location, and factors such as bandwidth and delay, providing practical insights for practitioners and researchers alike.

关键词： Performance evaluation Scheduling algorithms Soft sensors Scalability Bandwidth Logic gates Load management

来源：评论

学校读者我要写书评

暂无评论

An Image Analysis-Based Automated Method using Deep Learning for Grain Counting 17

An Image Analysis-Based Automated Method using Deep Learning...

引用

17th IEEE International Conference on Industrial and Information systems, ICIIS 2023

作者： Ajikaran, Ramesh Hewarathna, Ashen Iranga Palanisamy, Vigneshwaran Joseph, Charles Thuseethan, Selvarajah Sabaragamuwa University of Sri Lanka Dept. of Computing and Infromation Systems Sri Lanka Sabaragamuwa University of Sri Lanka Dept. of Software Engineering Sri Lanka Fau Dept. of Computer Science Germany School of Information Technology Deakin University Australia

ISBN: (纸本)9798350323634

In the process of quality analysis of grains, number counting is one of the key steps for agricultural production. Traditional manual grain counting methods are time-consuming and subject to human error, and automated methods have the potential to improve accuracy and save time. In this study, we aimed to develop an image analysis-based method to automatically quantify the number of grains in a quicker manner. The 576 grain images were collected manually, and labelImg tagging tool used to annotate to generate a text file with their respective positions by drawing bounding boxes manually. The datasets consist were separated into three groups: training, validation, and test. For object detection, the YOLOv5, YOLOv4, and YOLOv3 algorithms represent cutting-edge deep learning frameworks. They replace the tedious and error-prone manual counting process by precisely identifying and counting grains in images obtained from agricultural fields. This technique helps to increase grain counting's precision and effectiveness. We believe this method will be extremely beneficial in guiding the development of high throughput systems for counting the number of grains in other crops as it performs well with a wide range of backgrounds, picture sizes, grain sizes, as well as various quantities of grain crowding. When compared to the other two approaches, YOLOv4 performed well in terms of accuracy, speed, and robustness (97.65%), demonstrating that the suggested strategy is competitive with other cutting-edge deep networkst. © 2023 IEEE.

关键词： Grain (agricultural product)

来源：评论

学校读者我要写书评

暂无评论

RepCFT: A Fusion Model of Reparameterized Convolutional Neural Network and Visual Transformer for Wood Species Recognition 4

RepCFT: A Fusion Model of Reparameterized Convolutional Neur...

引用

4th International Conference on Artificial Intelligence and Computer engineering, ICAICE 2023

作者： Wang, Junyan Lian, Ziyan Kang, Feiyang Chen, Jinkai School of Internet Anhui University Hefei230039 China School of Software Engineering University of Science and Technology of China Hefei230026 China School of Advanced Technology University of Science and Technology of China Hefei230026 China Department of Network Management Center China Mobile Communications Group Anhui Co. LTD. Hefei230088 China

ISBN: (纸本)9798400708831

Wood is an extremely important foundational building material. Rapid and accurate wood species recognition is crucial for the rational application of wood and for preventing economic losses caused by the misidentification of wood species. Current research primarily focuses on image preprocessing and the use of Convolutional Neural Networks (CNNs) for wood species recognition. Most of the existing methods employ non-lightweight CNN models, making it difficult for the models to achieve rapid recognition. Addressing the above issues, this paper introduces a hybrid model that combines Reparameterized CNN and Visual Transformer, which we refer to as RepCFT. We have designed a new module, RepLG, with strong local and global information extraction capabilities. RepCFT, constructed from RepLG, achieves a recognition accuracy of 98.7% without any data augmentation or training strategies, while using only 4.3 million parameters and 0.6 billion Multiply-Accumulate operations (MACs). Its performance is on par with ResNet50 and surpasses some existing lightweight models. At the same time, it achieves a recognition speed of up to 601.46 frames per second, effectively meeting the need for rapid and accurate recognition in wood species recognition tasks. All code and models have been uploaded to https://***/OnlyForWW/RepCFT. © 2023 ACM.

关键词： Wood

来源：评论

学校读者我要写书评

暂无评论

CrossMeta: A Fast and Cheap Cross-Metaverse Interoperability Protocol

引用

IEEE Transactions on Services Computing 2025年

作者： Li, Taotao Yang, Qinglin Huang, Huawei Zhu, Xuanye Sun, Zhu Liu, Yuan Zheng, Zibin Sun Yat-Sen University School of Software Engineering Zhuhai campus Guangdong Zhuhai519082 China Guangzhou University Cyberspace Institute of Advanced Technology Huangpu campus Guangdong Guangzhou510006 China Singapore University of Technology and Design Singapore

Metaverse is drawing increasing attention from both academia and industry. Interoperability among different metaverse systems has become essential. A cross-metaverse interoperability protocol can enable interoperability across metaverses. However, cross-metaverse protocols often suffer significant cost overhead and transaction latency. For example, in STYLE, a leading cross-metaverse platform, 74% of transaction latency and 97% of the cost overhead are attributed to the relay blockchain rather than the two participating metaverses. To make cross-metaverse efficient, in this paper, we propose a fast and cheap cross-metaverse interoperability protocol named CrossMeta. CrossMeta can enable direct communication among heterogeneous metaverses rather than depending on a relay blockchain. This is achieved through two components: i) a committee that relays transactions from the source metaverse to the destination metaverse, along with availability proofs, and ii) a smart contract that verifies the proofs provided by the committee. To ensure an honest majority within the selected committee, we propose a dynamic committee selection method based on the chain quality property. Furthermore, we demonstrate that honest brokers achieve a Nash equilibrium. Additionally, we prove that the proposed CrossMeta protocol satisfies the security properties of atomicity and liveness. To demonstrate the practicality of CrossMeta, we implemented a prototype of the CrossMeta using two real-world metaverse platforms, i.e., Axie Infinity and Sandbox. The evaluation results show that CrossMeta outperforms other cross-metaverse solutions regarding transaction latency and gas fees. © 2008-2012 IEEE.

关键词： Blockchain

来源：评论

学校读者我要写书评

暂无评论

Current Status and Analyses of the SPLICE Hazard Detection software

Current Status and Analyses of the SPLICE Hazard Detection S...

引用

AIAA Science and technology Forum and Exposition, AIAA SciTech Forum 2025

作者： Liu, John Y. Chen, Po-Ting Aldrich, Jack D. Bocchino, Robert L. Vaughan, Geoffrey Jet Propulsion Laboratory California Institute of Technology Payload/Instrument Pointing Control Systems Engineering Guidance and Control Section 4800 Oak Grove Drive PasadenaCA91109 United States Jet Propulsion Laboratory California Institute of Technology Guidance and Control Section 4800 Oak Grove Drive PasadenaCA91109 United States Jet Propulsion Laboratory California Institute of Technology Flight Software and Avionics Systems Section 4800 Oak Grove Drive PasadenaCA91109 United States

This paper provides a status report of the current development and analysis efforts for the hazard detection (HD) software for the Safe and Precise Landing-Integrated Capabilities Evolution (SPLICE) project. © 20... 详细信息

ISBN: (数字)9781624107238

ISBN: (纸本)9781624107238

关键词： Hazards

来源：评论

学校读者我要写书评

暂无评论

FedCST: Federated Learning on Heterogeneous Resource-constrained Devices Using Clustering and Split Training 24

FedCST: Federated Learning on Heterogeneous Resource-constra...

引用

24th IEEE International Conference on software Quality, Reliability and Security Companion, QRS-C 2024

作者： Wang, Zhilu Lin, Haiyang Liu, Qi Zhang, Yonghong Liu, Xiaodong Nanjing University of Information Science and Technology School of Computer Science Jiangsu Nanjing China Jiangsu Province Engineering Research Center of Advanced Computing and Intelligent Services Jiangsu Nanjing China Nanjing University of Information Science and Technology School of Software Jiangsu Nanjing China Nanjing University of Information Science and Technology School of Automation Jiangsu Nanjing China Edinburgh Napier University School of Computing Edinburgh United Kingdom

ISBN: (纸本)9798350365658

With the rapid development of 5G and Internet of Things (IoT) technologies, edge devices such as sensors, smartphones, and wearable devices have become increasingly prevalent. The massive amount of distributed data generated by these devices offers unprecedented opportunities for deep learning, especially in fields like computer vision and edge computing. However, this data often exists in isolated data silos, and due to its privacy-sensitive nature, it cannot be directly processed in a centralized manner. Additionally, the data heterogeneity and devices heterogeneity further exacer-bates the challenges in federated learning (FL). Moreover, with the implementation of data privacy protection regulations, user demand for privacy protection is growing significantly. Therefore, how to effectively utilize dispersed data for distributed collaborative learning while ensuring data privacy is a pressing issue that needs to be addressed in the field of FL. In response to these challenges, this work has conducted research on clustering and segmentation FL algorithms for heterogeneous resource-constrained devices. A lightweight homogeneous device clustering strategy has been designed, which incorporates a split learning (SL) mechanism to enhance the accuracy and training efficiency of FL models. This approach reduces the load on resource-constrained devices and improves privacy security. Consequently, this method can meet the dual requirements of privacy protection and computational efficiency in real-world scenarios. © 2024 IEEE.

关键词： Federated learning

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：