检索结果-内蒙古大学图书馆

A novel dominating set and centrality based graph convolutional network for node classification

Multimedia Tools and Applications 2025年第14期84卷 12979-13004页

作者： Garg, Neeraj Garg, Sneha Jadeja, Mahipal Meena, Yogesh Kumar Gopalani, Dinesh Chauhan, Ganpat Singh Department of Computer Science and Engineering Malaviya National Institute of Technology Jaipur Rajasthan Jaipur302017 India Department of Computer Science and Engineering Swami Keshvanand Institute of Technology Jaipur Rajasthan Jaipur302017 India Department of Information Technology School of IT Manipal University Jaipur Rajasthan Jaipur302034 India

To learn and analyze graph-structured data, Graph Neural Networks (GNNs) have emerged as a powerful framework over traditional neural networks, which work well on grid-like or sequential structure data. GNNs are particularly designed to record dependencies and relations in large and complicated networks such as social networks, molecular structures, citation networks and recommendation systems. The main idea behind GNN is the message-passing mechanism that allows sharing of information between nodes and edges in a graph network, enabling the model to learn about graph structures and perform well even with limited training data. GNNs have three different architectures: Graph Convolutional Network (GCN), GraphSAGE, Graph Attention Model (GAT) and other hybrid models. In this paper, we have proposed novel GCN models namely dominating Set with Degree Centrality based GCN (DS_DC_GCN) and dominating Set with Betweenness Centrality based GCN (DS_BWC_GCN) model, for node classification task. In the traditional GNN models, training nodes are selected randomly. In our proposed models, we first identify a subset of the nodes with the help of a dominating set (D) of the underlying network. Then we consider final training nodes (say D′) from set D based on the centrality scores (degree or betweenness). We give preference to nodes having high centrality scores for training. To evaluate the performance of the proposed models, we have used three standard citation network datasets - Cora, Citeseer and Pubmed. Our models obtain improved node classification performance as compared to other state-of-the-art-methods. The consideration of high centrality nodes of dominating set in training helps to improve the feature availability in the underlying computational graphs, which in turn improves learning and hence, model performance. These models also have the potential to offer better results for other network tasks, including link prediction and community detection. © The Author(s), under

关键词： Graph neural networks

来源：评论

学校读者我要写书评

暂无评论

Analyzing Nickel Stock Price Prediction: A Comparative Study of Long Short-Term Memory and Convolutional Neural Network Models

Analyzing Nickel Stock Price Prediction: A Comparative Study...

引用

IEEE International Conference on Industry 4.0, Artificial Intelligence, and Communications Technology (IAICT)

作者： Ferry Irwanto Muhammad Farrell Hafizh Sasongko Stevanus Febrian Alexander Agung Santoso Gunawan Karli Eka Setiawan Data Science Program Computer Science Department School of Computer Science Bina Nusantara University Jakarta Indonesia Computer Science Department School of Computer Science Bina Nusantara University Jakarta Indonesia

ISBN: (数字)9798350353464

ISBN: (纸本)9798350353471

This research delves to predict PT Vale Indonesia Tbk stock price as an experiment on Indonesian stock using three models: naïve, LSTM, and 1D-CNN. Our analysis emphasizes the importance of matching model architectures to data properties. We compare models' performance with fiveday window for predict one-day prediction output. Interestingly, the single-layer LSTM outperforms the 1D-CNN even with similar hyperparameters, showcasing its strength in capturing long-term temporal dependencies crucial for nickel prices. While the 1D-CNN excels at identifying short-term patterns, its limited receptive field hinders long-term dependence. Recognizing the potential of both models, we encourage exploring hybrid architectures combining LSTM and CNN strengths for further improvement in financial forecasting. The experiment result shows single-layer LSTM outperforms a 1D-CNN with similar settings.

关键词： Analytical models Predictive models Nickel data models Communications technology Fourth Industrial Revolution Convolutional neural networks

来源：评论

学校读者我要写书评

暂无评论

Unveiling Veracity: An Integrated Semi-Supervised Algorithm for Fake News Detection 16

Unveiling Veracity: An Integrated Semi-Supervised Algorithm ...

引用

16th International Conference on Electronics, computers and Artificial Intelligence, ECAI 2024

作者： Malik, Aman Behera, Dayal Kumar Swain, Amulya Ratna Hota, Jhalak Birla Institute of Technology and Science Department of Computer Science & Information Systems Hyderabad India Kiit University School of Computer Engineering Bhubaneswar India

ISBN: (纸本)9798350371154

In recent years, the pervasive dissemination of misinformation and deliberately falsified content, commonly referred to as 'fake news,' has become a critical challenge in the realm of information dissemination. With the exponential growth of digital media and online platforms, the ease and speed at which misleading information can be generated and spread have increased substantially. Addressing the formidable task of detecting fake news has thus become an imperative within the field of information technology. The supervised model gives promising results. However, the availability of labelled data is critical in the realm of fake news detection. Leveraging unlabeled data alongside labelled data is an effective approach, but it requires careful handling. This paper explores the application of Semi-Supervised Learning for fake news detection by integrating Label Propagation and various ML models as base classifiers. By leveraging both labelled and unlabeled data, Label Propagation with CatBoost Classifier seeks to overcome traditional limitations, offering a promising avenue for bolstering the reliability of automated systems in distinguishing between authentic and deceptive information in the vast landscape of digital content. © 2024 IEEE.

关键词： Information dissemination

来源：评论

学校读者我要写书评

暂无评论

Classification of Health Insurance Fraud Risk with Machine Learning

Classification of Health Insurance Fraud Risk with Machine L...

引用

Information Technology Research and Innovation (ICITRI), International Conference on

作者： Muhammad Kent Al-Ghazi Ryan Bertrand Muhammad Dzul Qarrnayn Destra Alexander Agung Santoso Gunawan Karli Eka Setiawan Data Science Program Computer Science Department School of Computer Science Bina Nusantara University Jakarta Indonesia Computer Science Department School of Computer Science Bina Nusantara University Jakarta Indonesia

ISBN: (数字)9798350376210

ISBN: (纸本)9798350376227

Health insurance is a useful service that can help its users gain lifesaving medical aid when they are in need. However, health insurance is also exploitable to insurance fraud through the falsification of information to increase the amount of reimbursement and cause massive loss of funds to the insurance provider. We propose the usage of machine learning to accurately determine potential health insurance fraud. The objective of conducting this research is to determine which features are the most important to determine healthcare insurance fraud. This research used a dataset provided in Kaggle titled Healthcare Provider Fraud Detection Analysis using Random Forest Classifier and Logistic Regression. The best-performing model in this test, the Logistic Regression, is then used to which features are the most important for the classification. Our research shows that the most important feature in detecting health insurance fraud is the amount of money reimbursed associated with a provider. The Logistic Regression model achieved an accuracy of 0.90, precision of 0.93, recall of 0.91, and an F1 Score of 0.90, outperforming the Random Forest model in comparative analysis.

关键词： Logistic regression Analytical models Technological innovation Accuracy Insurance Medical services Feature extraction Fraud Information technology Random forests

来源：评论

学校读者我要写书评

暂无评论

Courier Delivery Optimization in Supply Chain via Group Multirole Assignment

Courier Delivery Optimization in Supply Chain via Group Mult...

引用

2024 IEEE International Conference on Systems, Man, and Cybernetics, SMC 2024

作者： Lin, Xuewei Ke, Xintong Zhu, Haibin Liu, Dongning School of Computer Science and Technology Guangdong University of Technology Guangzhou China Nipissing University Department of Computer Science and Mathematics North Bay Canada

ISBN: (纸本)9781665410205

Courier delivery is the end of the supply chain and affects the final delivery of products. Due to the promulgation of new courier delivery regulations, home delivery services have become the main choice for consumers, which has resulted in a greater workload. How to Reasonably Allocate Couriers for LEan Management (RACLEM) in supply chain optimization poses a challenge to traditional logistics. By extending the Group Multirole Assignment with Efficiency Degradation (GMRAED), this paper formalizes the problem. Moreover, we propose a quantitative calculation method of efficiency degradation based on Amdahl's law, taking region similarity and the number of tasks as important criteria, and compare the performance of different personnel arrangements. Additionally, compared with the brute force algorithm, we combine GMRAED and Genetic Algorithm (GA) to create a practical solution. Large-scale simulation experiments demonstrate that the genetic algorithm significantly shortens the solution time, and the lowest experimental accuracy is 99.407%. By using the above method, decision-makers are able to assist companies optimize personnel resource allocation, minimize personnel waste, and improve the performance of courier delivery within a shorter timeframe. © 2024 IEEE.

关键词： Efficiency

来源：评论

学校读者我要写书评

暂无评论

ESOA-Based 1D-CNN Classification for Amazon Textual Review Sentiment Analysis of Mobile Phones

ESOA-Based 1D-CNN Classification for Amazon Textual Review S...

引用

International Conference on Computing and Machine Learning, CML 2024

作者： Kosala, N. Nirmalrani, V. School of Computing Department of Computer Science and Engineering Sathyabama Institute of Science and Technology Tamilnadu Chennai India

ISBN: (纸本)9789819775484

A trend toward the use of digital platforms for product marketing and profit margin enhancement is evident in the contemporary corporate environment. As a result, using deep learning methods to evaluate consumer sentiment in tweets based on product reviews becomes essential. The study's proposal is to use a big dataset and a one-dimensional convolutional neural network (1D-CNN) for the classification process in order to extract lexical characteristics for sentiment analysis. In order to do this, analysis is done on customer review datasets of Amazon items with an emphasis on reviews pertaining to mobile phones. To improve the performance and accuracy of the classifier, data preprocessing is first done. A lexical method is then used to efficiently compress the text via feature extraction. For the purpose of correctly assessing review emotions, efficient classification is essential. So, classification is done using the 1D-CNN method. The Egret Swarm Optimization Algorithm is used to adjust the 1D-CNN model's hyperparameters (ESOA). After this procedure, the research evaluates the 1D-CNN method's performance in relation to other techniques currently in use. The research concludes that the suggested method performs well overall, achieving 98.77% accuracy, 97.37% precision, 97.42% sensitivity, and 97.21% F-measure, respectively. These performance measures outperform those of current algorithms created in earlier research. The sentiment analysis of mobile phone evaluations inside the Amazon e-commerce system shows improved accuracy and efficiency when using the suggested 1D-CNN neural network approach. © The Author(s), under exclusive license to Springer Nature Singapore Pte Ltd. 2024.

关键词： Convolutional neural networks

来源：评论

学校读者我要写书评

暂无评论

Accelerating local SGD for non-IID data using variance reduction

引用

Frontiers of computer science 2023年第2期17卷 73-89页

作者： Xianfeng LIANG Shuheng SHEN Enhong CHEN Jinchang LIU Qi LIU Yifei CHENG Zhen PAN Anhui Province Key Laboratory of Big Data Analysis and Application University of Science and Technology of ChinaHefei 230027China Ant Financial Services Group Hangzhou 310000China Department of Computer Science and Engineering Hong Kong University of Science and TechnologyHong Kong 999077China

Distributed stochastic gradient descent and its variants have been widely adopted in the training of machine learning models,which apply multiple workers in *** them,local-based algorithms,including Local SGD and FedAvg,have gained much attention due to their superior properties,such as low communication cost and ***,when the data distribution on workers is non-identical,local-based algorithms would encounter a significant degradation in the convergence *** this paper,we propose Variance Reduced Local SGD(VRL-SGD)to deal with the heterogeneous *** extra communication cost,VRL-SGD can reduce the gradient variance among workers caused by the heterogeneous data,and thus it prevents local-based algorithms from slow convergence ***,we present VRL-SGD-W with an effectivewarm-up mechanism for the scenarios,where the data among workers are quite *** from eliminating the impact of such heterogeneous data,we theoretically prove that VRL-SGD achieves a linear iteration speedup with lower communication complexity even if workers access non-identical *** conduct experiments on three machine learning *** experimental results demonstrate that VRL-SGD performs impressively better than Local SGD for the heterogeneous data and VRL-SGD-W is much robust under high data variance among workers.

关键词： distributed optimization variance reduction local SGD federated learning non-IID data

来源：评论

学校读者我要写书评

暂无评论

Detection of Plant Disease Using Machine Learning and Deep Learning Algorithms 8

Detection of Plant Disease Using Machine Learning and Deep L...

引用

8th IEEE International Conference on science, Technology, Engineering and Mathematics, ICONSTEM 2023

作者： Gollapudi, Kalyani Varigonda, Preethi Jancy, S. Pushpalatha, S. Marya, Viji Amutha Helen, Suji Sathyabama Institute of Science and Technology Computer Science and Engineering Chennai India Sathyabama Institute of Science and Technology Department of Computer Science and Engineering Chennai India Saveetha School of Engineering Simats Department of Data Analytics Chennai602105 India

ISBN: (纸本)9798350347791

Agriculture performs an critical position in India's economic system. Early detection of plant illnesses is critical to save you crop damage and similarly spread of diseases. Most plants, along with apple, tomato, cherry, grape, show symptoms of leaf ailment. These visible patterns may be found to correctly predict the disorder and take early movement to save you it. This can be triumph over with system getting to know and deep getting to know algorithms. We therefore recommend a method that determines tomato plant disease from pix of leaves. This method is performed with aid vector device (SVM), random woodland gadget studying algorithm, and look at algorithms Convolutional Neural Network (CNN), Recurrent Neural Network (RNN), and ResNet that is one of the switch learning techniques. Snoring. After the facts set is processed by way of the algorithms, the accuracy of the algorithms is compared and the snapshots are categorized. © 2023 IEEE.

关键词： Recurrent neural networks

来源：评论

学校读者我要写书评

暂无评论

Usability Analysis of Text Generation by ChatGPT OpenAI Using System Usability Scale Method 8

Usability Analysis of Text Generation by ChatGPT OpenAI Usin...

引用

8th International Conference on computer science and Computational Intelligence, ICCSCI 2023

作者： Mulia, Angelina Patience Piri, Pirelli Rahelya Tho, Cuk Computer Science Department School of Computer Science Bina Nusantara University Malang65154 Indonesia Computer Science Department Bina Nusantara University Jakarta11480 Indonesia

The development of artificial intelligence systems has resulted in various AI products including ChatGPT, which is a new product classified as a chatbot. This research aims to ensure that text generation systems such as ChatGPT open AI have the best level of quality and usability and are able to provide a satisfying experience for users. To measure and evaluate the effectiveness, efficiency and user satisfaction of the ChatGPT platform, researchers used the System Usability Scale (SUS) method. This data collection was carried out using an online questionnaire. After the collected data has been tested for validity and reliability, the researchers then analyzed the data results. From the results of the research conducted, the SUS value of the ChatGPT platform is 67.44. This score is included in the marginal high category of class D, with a reasonable or sufficient interpretation. With the results of the analysis per question item, it shows that users tend to agree that the system runs quite effectively, efficiently, well and is easy to understand. Although ChatGPT is able to perform tasks or commands well. However, it should be noted that not all information loaded by ChatGPT is presented in a complete, current and correct manner. This is because the information presented by ChatGPT is only limited to 2021. Because ChatGPT is a new technology and is still under development, further researchers are expected to test other features or ChatGPT to ensure the stability and reliability of the entire ChatGPT system using other research methods. © 2023 The Authors. Published by Elsevier B.V.

关键词： Artificial Intelligence ChatGPT OpenAI Questionnaire SUS System Usability Scale Text Generation Usability Analysis

来源：评论

学校读者我要写书评

暂无评论

A hyperledger fabric-based EMR sharing mechanisms with proxy re-encryption and IPFS

引用

Cluster Computing 2024年第5期27卷 6953-6983页

作者： Huang, Der-Chen Liu, Ling-Chun Deng, Yong-Yuan Chen, Chin-Ling Zeng, Kuang-Wei Department of Computer Science and Engineering National Chung-Hsing University Taichung402 Taiwan School of Information Engineering Changchun Sci-Tech University Jilin Changchun130600 China Department of Computer Science and Information Engineering Chaoyang University of Technology Taichung41349 Taiwan

With the advent of the information age, data storage has not only developed from paper information systems to electronic information system storage, but has also extended to cloud database storage methods. To date, we can see the application of big data and cloud in various fields and levels, making induction, arrangement, transmission, and exchange of data easier and faster. The electronic and systematisation of patient medical records in the medical environment is an example. The use of electronic medical records can help improve the quality of medical care, reduce duplication of treatment, and reduce the waste of examination resources. It has many advantages, but also many risks. Medical institutions may abuse electronic medical data without the consent of patients, leading to the leakage of patient privacy data and causing serious social impact. Therefore, this study proposes a medical record exchange and sharing mechanism based on the Hyperledger blockchain architecture combined with proxy reencryption. When pharmacists need to obtain medical records for related research, patients can decide whether to share medical records, and hospitals can also revoke pharmacist access rights at any time to achieve more private, secure, and convenient sharing of medical records. The proposed scheme meets various blockchain security requirements and the BAN (Burrows–Abadi–Needham) logic proof model (Burrows–Abadi–Needham) is applied to assess the correctness of the proposed scheme. The proposed scheme performs well in terms of computational and communication costs. © The Author(s), under exclusive licence to Springer science+Business Media, LLC, part of Springer Nature 2024.

关键词： Blockchain

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：