检索结果-内蒙古大学图书馆

7th International Conference on Smart Computing and Communication, SmartCom 2022

作者： Xiao, Jiahong Liang, Wei Cai, Jiahong Zhu, Hangyu Li, Xiong Xie, Songyou School of Computer Science and Engineering Hunan University of Science and Technology Xiangtan411201 China Hunan Key Laboratory for Service Computing and Novel Software Technology Xiangtan411201 China Guangdong Financial High-Tech Zone "Blockchain +" Fintech Research Institute Foshan528253 China School of Computer Science and Engineering University of Electronic Science and Technology of China Chengdu611731 China College of Computer Science and Electronic Engineering Hunan University Changsha410082 China

ISBN: (纸本)9783031281235

Nowadays, blockchain distributed ledger technology is becoming more and more prominent, and its decentralization, anonymization, and tampering obvious features have been widely recognized. These excellent technical features of blockchain have also made it a hot issue for global research. With the wide application of blockchain technology in various industries, some defects are gradually exposed, and more prominently, the blockchain system is unable to meet the current demand of explosive growth of data volume and frequent data interaction. As one of the key technologies to solve this problem, sharding technology is gaining attention. This article introduces common blockchain scaling schemes and focuses on an overview of blockchain sharding. Sharding technology is introduced from two perspectives of intra-slice consensus and inter-slice consensus. The current mainstream slicing technology is summarized according to three different slicing methods: network sharding, transaction sharding, and state sharding. Finally, the challenges faced by current blockchain sharding technology are analyzed and the full text is summarized. © 2023, The Author(s), under exclusive license to Springer Nature Switzerland AG.

关键词： Blockchain

来源：评论

学校读者我要写书评

暂无评论

Using LLMs for Annotation and ML Methods for Comparative Analysis of Large-Scale Turkish Sentiment Datasets

Using LLMs for Annotation and ML Methods for Comparative Ana...

引用

International Conference on computer Science and engineering (UBMK)

作者： Ercan Ezin Rukiye Savran Kiziltepe Murat Karakus Dept. of Computer Engineering Harran University Şanlıurfa Türkiye Dept. of Software Engineering Karadeniz Tech. University Trabzon Türkiye Dept. of Software Engineering Ankara University Ankara Türkiye

ISBN: (数字)9798350365887

ISBN: (纸本)9798350365894

With the increasing prevalence of e-commerce plat-forms, understanding customer sentiments expressed in product reviews is crucial for assessing platform and product performance. However, traditional sentiment analysis (SA) models often struggle with low-resource languages such as Turkish, particularly due to the language's agglutinative nature and morphological complexity. To address this, several public datasets are now available online. However, these datasets are often annotated using rule-based methods, resulting in poor accuracy for Machine Learning (ML) models. This study addresses these challenges by employing two large Turkish datasets, TRSAv1 and VSCR, and improving their sentiment annotations through advanced Large Language Models (LLMs) such as ChatGPT4o-mini. The reannotation significantly enhanced the performance of traditional ML models-Logistic Regression, Naive Bayes, Random Forest, and Support Vector Machine-especially in the neutral and negative sentiment classes, as evidenced by increased precision, recall, and F1-scores. The findings underscore the importance of thorough data preparation and demonstrate that utilizing LLMs for sentiment annotation in Turkish can lead to more reliable SA, providing a stronger foundation for developing advanced SA tailored to low-resource languages. This study is the first to employ recent LLM for relabelling Turkish customer reviews, highlighting the potential for improved accuracy in SA for low-resource languages.

关键词： Analytical models Accuracy Systematics Annotations Reviews Computational modeling Transfer learning Transformers Vectors Electronic commerce

来源：评论

学校读者我要写书评

暂无评论

Probabilistic Graph Modeling based Safety Classifier Algorithm for Smart Transportation

Probabilistic Graph Modeling based Safety Classifier Algorit...

引用

2023 International Symposium on Green Technologies and Applications, ISGTA 2023

作者： Karkouri, Najib El Tigani, Smail Saadane, Rachid Chehri, Abdelah Pierre, Samuel Neya, Noureddine Mathematics and Informatics Lab Faculty of Sciences University Ibno Toufial Kenitra Morocco Research and Development Unit Accsellium LLC Fez Morocco SIRC-LaGeS Hassania School of Public Labors Casablanca20250 Morocco Department of Mathematics and Computer Science Royal Military College of Canada Kingston Canada Department of Computer and Software Engineering Polytechnique Montreal Canada

This research introduces a novel Probabilistic Graph Modeling-based Safety Classifier Algorithm designed for the purpose of classifying road safety in smart transportation systems. Leveraging a combination of numerical integration methods and Gaussian kernel density estimation models, the proposed algorithm offers an effective approach to assess and categorizing the safety levels of roads. The integration of these techniques enables a comprehensive analysis, allowing for a more nuanced understanding of the complex interactions within transportation networks. The algorithm's efficacy in road safety classification holds promising implications for enhancing traffic management and promoting safer urban environments. © 2024 The Authors. Published by ELSEVIER B.V.

关键词： Accident prevention

来源：评论

学校读者我要写书评

暂无评论

Combinatorial Incentive Mechanism for Bundling Spatial Crowdsourcing with Unknown Utilities

Combinatorial Incentive Mechanism for Bundling Spatial Crowd...

引用

IEEE Annual Joint Conference: INFOCOM, IEEE computer and Communications Societies

作者： Hengzhi Wang Laizhong Cui Lei Zhang Linfeng Shen Long Chen College of Computer Science and Software Engineering Shenzhen University China School of Computing Science Simon Fraser University Canada

ISBN: (数字)9798350383508

ISBN: (纸本)9798350383515

Incentive mechanisms in Spatial Crowdsourcing (SC) have been widely studied as they provide an effective way to motivate mobile workers to perform spatial tasks. Yet, most existing mechanisms only involve single tasks, neglecting the presence of complementarity and substitutability among tasks. This limits their effectiveness in practice cases. Motivated by this, we consider task bundles for incentive mechanism design and closely analyze the mutual exclusion effect that arises with task bundles. We then develop a combinatorial incentive mechanism, including three key policies: In the offline case, we propose a combinatorial assignment policy to address the conflict between mutual exclusion and assignment efficiency. We next study the conflict between mutual exclusion and truthfulness, and build a combinatorial pricing policy to pay winners that yields both incentive compatibility and individual rationality. In the online case with unknown workers’ utilities, we present an online combinatorial assignment policy that balances the exploration-exploitation trade-off under the mutual exclusion constraints. Through theoretical analysis and numerical simulations using real-world mobile networking datasets, we demonstrate the effectiveness of the proposed mechanism.

关键词： Crowdsourcing Mechanism design Pricing Numerical simulation Task analysis

来源：评论

学校读者我要写书评

暂无评论

FPGNet: Single Image Deraining with High-Frequency Channel and Frequency Domain Prior Guidance

FPGNet: Single Image Deraining with High-Frequency Channel a...

引用

International Conference on Acoustics, Speech, and Signal Processing (ICASSP)

作者： Zhaoyong Yan Gang Li Lingyu Si Hongwei Dong School of Computer Engineering and Science Shanghai University Institute of Software Chinese Academy of Sciences University of Chinese Academy of Sciences

In recent years, deep learning methods have shown promising results in Single Image Deraining (SID). However, these methods still suffer from unsatisfactory residual rain streaks, primarily due to the absence of image priors embedding and limitations in modeling capacity. In this paper, we propose a joint High-Frequency Channel and Fourier Frequency Domain Guided Network (FPGNet) to remove complex rain streaks while obtaining clearer background details. Specifically, FPGNet comprises two core components: the High-frequency Guidance Module (HFGM) and the Fourier-based Multi-scale Feature Extraction Module (FMFEM). The HFGM focuses on learning the natural distribution of rain streaks from the high-frequency channel to guide the network’s modeling of rain streaks, while the FMFEM serves as the backbone of FPGNet to learn rain layers in rainy images. Additionally, we introduce an Interactive Fusion Module (IFM) to enhance the integration of the high-frequency channel into the network. Extensive experiments demonstrate that our network outperforms state-of-the-art deraining methods in effectively removing rain streaks from rainy images.

关键词：

来源：评论

学校读者我要写书评

暂无评论

A New End-to-End Image Contrastive Clustering Based on Mining Latent Positive Samples

A New End-to-End Image Contrastive Clustering Based on Minin...

引用

Machine Learning and Natural Language Processing (MLNLP), International Conference on

作者： Bo Feng Xi Yan Li Yan Bing Luo Zheng Pei School of science Xihua University Chengdu China School of Computer Science and Software Engineering Xihua University Chengdu China

ISBN: (数字)9798350354973

ISBN: (纸本)9798350354980

The integration of the contrastive learning paradigm into deep clustering has led to enhanced performance in image clustering. However, in existing researches, the samples in the class of the target may be still treated as negative samples, this means that more semantic sample pairs cannot be constructed and the intra-class compactness may be compromised. In this paper, we propose a new contrastive clustering method based on mining latent positive samples (CCLPS), which utilizes the nearest neighbor relationship between samples to mine semantic positive samples in the same class. In more detail, by mining latent positive samples in both the feature and cluster spaces, CCLPS can obtain more cluster-friendly feature representation of samples to enhance clustering performance. Samples mined from the feature space fuel the cluster-level contrastive learning task, while samples mined from the cluster space fuel the instance-level task. Three image datasets CIFAR-10, CIFAR-100 and Tiny-ImageNet are employed to experimentally demonstrate usefulness and effectiveness of the proposed method in image clustering, and transfer learning based on mining latent positive samples is also discussed and comparisons are made with 19 representative clustering methods on the datasets. Experimental results indicate that the proposed method can serve as a pre-training model for transfer learning and exhibits commendable feature modeling capabilities.

关键词： Training Clustering methods Transfer learning Semantics Contrastive learning Benchmark testing Natural language processing Fuels Standards

来源：评论

学校读者我要写书评

暂无评论

Dense Error Map Estimation for MRI-Ultrasound Registration in Brain Tumor Surgery Using Swin UNETR

arXiv

引用

arXiv 2023年

作者： Salari, Soorena Rasoulian, Amirhossein Rivaz, Hassan Xiao, Yiming Department of Computer Science and Software Engineering Concordia University MontrealQC Canada Department of Electrical and Computer Engineering Concordia University MontrealQC Canada

Early surgical treatment of brain tumors is crucial in reducing patient mortality rates. However, brain tissue deformation (called brain shift) occurs during the surgery, rendering pre-operative images invalid. As a cost-effective and portable tool, intra-operative ultrasound (iUS) can track brain shift, and accurate MRI-iUS registration techniques can update pre-surgical plans and facilitate the interpretation of iUS. This can boost surgical safety and outcomes by maximizing tumor removal while avoiding eloquent regions. However, manual assessment of MRI-iUS registration results in real-time is difficult and prone to errors due to the 3D nature of the data. Automatic algorithms that can quantify the quality of inter-modal medical image registration outcomes can be highly beneficial. Therefore, we propose a novel deep-learning (DL) based framework with the Swin UNETR to automatically assess 3D-patch-wise dense error maps for MRI-iUS registration in iUS-guided brain tumor resection and show its performance with real clinical data for the first time. © 2023, CC BY-NC-ND.

关键词： Ultrasonics

来源：评论

学校读者我要写书评

暂无评论

UAV-Assisted Maritime Data Collection via Optical Communications using Deep Reinforcement Learning

UAV-Assisted Maritime Data Collection via Optical Communicat...

引用

International Conference on Big Data Computing and Communications (BIGCOM)

作者： Saisai Ma Mengzhen Li Tinglong Deng Jinglong Wang Rukhsana Ruby Hanjiang Luo College of Computer Science and Engineering Shandong University of Science and Technology Qingdao China College of Computer Science and Software Engineering Shenzhen University Shenzhen China

Ocean data collection via unmanned aerial vehicles (UAVs) has received widespread attention due to its flexibility and low cost. To further improve the efficiency of large maritime data collection between the UAVs and the ocean surface buoys, optical wireless transmission is considered as a promising technique because of its low latency and high bandwidth. However, the waves and other disturbances in the complex oceanic environment result in drift and instability of the buoy, which deteriorate and even interrupt the line-of-sight (LOS) optical transmission. To tackle these challenges, in this paper, we propose a reliable data collection scheme based on a deep reinforcement learning (DRL) approach. We first model the optical channel and calculate the maximum transmission range that satisfies a pre-defined bit error rate (BER) to ensure the quality of service (QoS). Then, we formulate the data collection procedure as a Markov decision process (MDP) aiming at maximizing the received signal intensity and minimizing the energy consumption, in which the system uncertainty caused by the oceanic environmental disturbances is taken into account. Finally, we propose a beam pointing adjustment algorithm based on deep deterministic policy gradient (DDPG) approach to alleviate the performance deterioration and maintain a stable LOS communication. Through extensive simulations, the results demonstrate that the proposed scheme is effective and achieves reliable data collection via the optical links.

关键词：

来源：评论

学校读者我要写书评

暂无评论

Parallel Multi-objective Multi-type Robot Disassembly Line Balancing Problems

Parallel Multi-objective Multi-type Robot Disassembly Line B...

引用

2023 International Conference on Cyber-Physical Social Intelligence, ICCSI 2023

作者： Zhang, Zhiwei Guo, XiWang Wang, Jiacun Qin, Shujin Qi, Liang Tang, Ying Information and Control Liaoning Petrochemical University Fushun China Information and Control Liaoning Petrochemical University Monmouth University United States Monmouth University West Long Branch Computer Science and Software Engineering United States Economics and Management Shangqiu Normal University Shangqiu China Shandong University of Science and Technology Computer Science and Technology Qingdao China Roman University Electrical and Computer Engineering Glassboro United States

ISBN: (纸本)9798350312492

Remanufacturing of waste products cannot be separated from disassembly. Reasonable disassembly sequence can realize balanced production of disassembly line, increase work efficiency during remanufacturing, reduce capital investment, reduce pollutant emission, improve safety level and increase parts utilization rate. In previous studies, few have considered the balance between profit and the environment. To tackle this problem, the paper proposes an enhanced parallel layout disassembly line model and utilizes an improved multi-objective multi-verse optimization approach to obtain superior solutions. The accuracy of the model and the effectiveness of the proposed algorithm are validated by comparing the results with those obtained using other algorithms. © 2023 IEEE.

关键词： Multiobjective optimization

来源：评论

学校读者我要写书评

暂无评论

Triple Diseases Prediction System using Deep Learning Techniques

Triple Diseases Prediction System using Deep Learning Techni...

引用

Emerging Trends in Information Technology and engineering (ic-ETITE), Conference on

作者： R. Vijayan V. Mareeswari A. B. Jaswanth Department of Information Technology School of Computer Science Engineering and Information Systems (SCORE) Vellore Institute of Technology (VIT) Vellore Tamilnadu Department of Software and Systems School of Computer Science Engineering and Information Systems (SCORE) Vellore Institute of Technology (VIT) Vellore Tamilnadu MTech (Software Engineering) School of Computer Science Engineering and Information Systems (SCORE) Vellore Institute of Technology (VIT) Vellore Tamilnadu

With the existing deep learning models in predicting multiple diseases primarily focus on analyzing individual diseases in isolation, lacking a unified system for multi-disease prediction. This project presents an approach to predict multiple diseases using Flask API, with a specific focus on brain tumors, COVID-19 and pneumonia. The proposed work represents a significant contribution to the field of disease prediction, harnessing the power of deep learning algorithms and modern web application development. The primary focus is on disease prediction, with a particular emphasis on ensuring accuracy and accessibility for end-users. The initial phase of this research involves data collection, where relevant datasets of various diseases are gathered. These datasets serve as the foundation for training and validating the deep learning models. Two prominent deep learning algorithms, Sequential CNN and VGG16, are employed for this purpose. These algorithms are chosen for their ability to handle complex data and recognize patterns within medical images and other health-related data. The core of the research involves training the deep learning models using the collected datasets. This step is crucial in enabling the models to learn and generalize from the provided data, ultimately enhancing their predictive capabilities. The models are modified to elevate their performance and accuracy. Following the training phase, the models are rigorously tested to evaluate their predictive accuracy. This assessment is vital in gauging the real-world applicability of the models in medical diagnosis. To make these powerful disease prediction models accessible to a wider audience, a front-end web application is developed.

关键词：

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：