检索结果-内蒙古大学图书馆

Risk warning based on GA-improved stacking ensemble learning algorithm: a case study of alcohol

Annals of Operations Research 2025年 1-28页

作者： Song, Cen Shen, Hanwen Zhuang, Jun School of Economics and Management China University of Petroleum No. 18 Fuxue Road Changping District Beijing 102249 China Department of Industrial and Systems Engineering University at Buffalo Buffalo 14260 NY United States

As one of the important daily consumer goods, alcoholic beverage presents high safety risks and potential hazards. Therefore, ensuring its quality and safety to meet consumer demand is urgent. Meanwhile, in response to the iterative and optimization issues in ensemble learning, this paper introduces a Genetic Algorithm (GA) to enhance the performance of the Stacking ensemble learning model. Based on the inspection data of alcoholic food products published by National Food Safety Sampling Inspection Results Query System in China, this study applies the Synthetic Minority Oversampling Technique and Tomek Links algorithm for comprehensive sampling, to achieve balanced categorization of different samples. It effectively mitigates the impact of non-smooth data and improves the classification results biased towards majority class samples. After cross-validation and hyperparameter optimization, the proposed GA-improved Stacking ensemble learning model has an accuracy of 0.885, a precision of 0.882, a recall of 0.885, and an F1 score of 0.876, comparing with traditional single classifier algorithms such as random search, Bayesian search, and other methods. This study provides an effective risk warning method for regulatory agencies' inspection focus. © The Author(s), under exclusive licence to Springer Science+Business Media, LLC, part of Springer Nature 2025.

关键词： Ensemble learning Genetic algorithm Risk warning Single classifier algorithm

来源：评论

学校读者我要写书评

暂无评论

Optimizing Coverage in Wireless Sensor Networks Using the Tumbleweed Algorithm

Journal of Network Intelligence

引用

Journal of Network Intelligence 2025年第1期10卷 250-261页

作者： Zhang, Mengfei Chu, Shu-Chuan Liu, Nengxian Huang, Kuan-Chun Pan, Jeng-Shyang College of Computer Science and Engineering Shandong University of Science and Technology Qingdao266590 China College of Computer and Data Science Fuzhou University Fuzhou350108 China Department of Industrial and Systems Engineering Chung Yuan Christian University Taoyuan320314 Taiwan School of Artificial Intelligence Nanjing University of Information Science and Technology Nanjing210044 China Department of Information Management Chaoyang University of Technology Taichung41349 Taiwan

Wireless sensor networks (WSNs) have found extensive applications across various fields, significantly enhancing the convenience in our daily lives. Hence, an in-creasing number of researchers are directing their attention toward the performance of sensor nodes, with a particular emphasis on addressing the coverage challenges in WSNs. Node coverage plays a crucial role in determining the detection performance of WSNs, making it imperative to attain maximum coverage with a given number of sensor nodes. This paper applies the Tumbleweed Algorithm (TA) to the coverage problem in WSNs and validates its effectiveness through simulation experiments. The results demonstrate its unique advantages in global search and avoiding local optima, providing a new solution for the WSN coverage problem and offering valuable references for future research and practical applications. The TA algorithm simulates the lifecycle of tumbleweed from seedling to mature plant and includes the concept of growth cycles. This paper compares the TA algorithm with PSO, GWO, DE, BOA, SCA, and AOA, setting different numbers of wireless sensor nodes and varying sensing radii for the experiments. The experimental results show that the TA algorithm significantly improves coverage rates under various configurations and consistently outperforms other commonly used algorithms. © 2025, Taiwan Ubiquitous Information CO LTD. All rights reserved.

关键词： Sensor nodes

来源：评论

学校读者我要写书评

暂无评论

Towards AI-enhanced process planning: Assessing machine tool capability based on part design 6

Towards AI-enhanced process planning: Assessing machine tool...

引用

6th International Conference on Industry 4.0 and Smart Manufacturing, ISM 2024

作者： Abolghasem, Sepideh Youssef, Matthew Abedrabbo, Faruk Pandde, Amman Department of Manufacturing Systems Engineering and Management California State University NorthridgeCA91330 United States Faculty of Technical Sciences International University of Ecuador UIDE Quito170411 Ecuador

The emergence of the fourth industrial revolution, or Industry 4.0, necessitates a more automated approach to manufacturing process planning. This process begins with evaluating machine tool capabilities to handle specific part geometries and microstructures. Once a match is established, the focus shifts to developing an efficient method for converting design elements into physical components. This work aims to create and validate a framework that assesses the manufacturability of design features based on the available machinery and materials. Specifically, it involves classifying manufacturing processes, such as turning and milling, for a given part design geometry. To achieve this, feature attributes like rotational symmetry and D2 distribution are calculated for a dataset used to train a decision tree. This model then suggests the appropriate manufacturing process for a given CAD model. The decision tree is validated with a separate dataset, showing reasonable accuracy. Ultimately, the goal is to enhance process planning, ensuring the seamless translation of designs into physical products, with a particular emphasis on geometry, microstructure, and cost. © 2025 The Authors. Published by Elsevier B.V.

关键词： Design for manufacturability

来源：评论

学校读者我要写书评

暂无评论

Integrating Potential Pronunciations for Enhanced Mispronunciation Detection and Diagnosis Ability in LLMs

Integrating Potential Pronunciations for Enhanced Mispronunc...

引用

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

作者： Minglin Wu Jing Xu Xueyuan Chen Helen Meng Department of Systems Engineering and Engineering Management The Chinese University of Hong Kong Hong Kong SAR China

ISBN: (数字)9798350368741

ISBN: (纸本)9798350368758

Large Language Models (LLMs) have exhibited significant potentials across various tasks. However, how to leverage the power of LLMs in the mispronunciation detection and diagnosis (MDD) task is still under-explored. In this paper, we propose a PP-ATP model, which integrates potential pronunciations covering common mispronunciations into the prompt part of LLMs, to enhance the MDD ability of LLMs in second language (L2) English. Specifically, the proposed PP-ATP model is composed of an audio encoder, an LLM decoder, and an adapter. Taking speech representations from the audio encoder as the audio prompt and reference sentence with canonical and potential pronunciations as text prompt, the LLM decoder is adapted to predict the actual pronunciation in the given L2 speech. Experiments show that our PP-ATP model achieves new state-of-the-art (SOTA) performance in MDD on CU-CHLOE corpus, confirming the effectiveness of potential pronunciation integration.

关键词： Adaptation models Large language models Signal processing Linguistics Acoustics Decoding Speech processing

来源：评论

学校读者我要写书评

暂无评论

A Critique on Transductive Evaluation for GNN Node Classification 12th

A Critique on Transductive Evaluation for GNN Node Classif...

引用

12th International Symposium on From Data Models and Back, DataMod 2024

作者： Guo, Yameng Vanden Broucke, Seppe Department of Business Informatics and Operations Management Ghent University Tweekerkenstraat 2 Ghent9000 Belgium Research Centre for Information Systems Engineering KU Leuven Naamsestraat 69 Leuven Belgium

ISBN: (纸本)9783031879074

Transductive learning and inductive learning are two standard paradigms widely used in deep learning for graph data. Transductive learning focuses on inference on seen data by incorporating features of the complete dataset into the training phase, while inductive learning infers on unseen instances and has the capability of generalization. The differences in learning goals require distinct data partitioning approaches, where the data is typically divided into training and test sets. Unlike tabular data, the division of graph data is constrained by its internal topology, making it difficult to obtain a clear-cut test set when considering only a single graph. As a result, most research adopts a transductive setting for graph learning and applies a masking strategy during training, which masks labels but allows the use of graph attributes, including those related to the masked entities. We argue that transductive learning and evaluation in Graph Neural Networks (GNNs) tasks, particularly for node classification, suppresses the potential of GNNs to generalize. As an alternative, we propose a data splitting approach for inductive learning to train and evaluate GNN performance for graph datasets consisting of a single graph. We demonstrate the feasibility of inductive evaluation for both training and inference on unseen data within a complete graph by means of an experiment on well-known benchmark graph datasets. © The Author(s), under exclusive license to Springer Nature Switzerland AG 2025.

关键词： Deep learning

来源：评论

学校读者我要写书评

暂无评论

Endothelial dysfunction promotes age-related reorganization of collagen fibers and alters aortic biomechanics in mice

引用

American Journal of Physiology - Heart and Circulatory Physiology 2025年第4期328卷 H900-H914页

作者： Du, Liya Rodgers, Jeffrey Gharraee, Nazli Gary, Olivia Shazly, Tarek Eberth, John F. Lessner, Susan M. Biomedical Engineering Program College of Engineering and Computing University of South Carolina Columbia SC United States College of Arts and Sciences University of South Carolina Columbia SC United States Department of Cell Biology and Anatomy School of Medicine University of South Carolina Columbia SC United States Biomedical Engineering Science and Health Systems Drexel University Philadelphia PA United States

Endothelial dysfunction, defined as a reduction in the bioavailability of nitric oxide (NO), is a risk factor for the occurrence and progression of various vascular diseases. This study investigates the effect of endothelial dysfunction on age-related changes in aortic extracellular matrix (ECM) microstructure and the relationship between microstructural adaptation and the mechanical response. Here, we used groups of NOS3 knockout (KO), NOS3 heterozygotes (Het), and wild-type (WT) B6 mice (controls) to study changes in hemodynamic parameters, collagen fiber organization, and both active and passive aortic mechanics using biaxial pressure myography over a time course from 1.5 to 12 mo. Our results show that homeostatic levels of passive circumferential stress and stretch were preserved in KO mice by remodeling adventitial collagen fibers toward a more predominantly circumferential direction with age, rather than by increased fibrosis, in response to hypertension induced by endothelial dysfunction. However, passive aortic stiffness in KO mice was significantly increased owing to geometrical changes, including significant increases in wall thickness and decreases in inner diameter, and by ECM microstructural reorganization, during this maladaptive vascular remodeling. Furthermore, long-term NO deficiency significantly increased smooth muscle cell (SMC) contractility initially, but this effect was attenuated with age. These findings improve our understanding of microstructural and mechanical changes during the maladaptive vascular remodeling process, demonstrating a role for adventitial collagen fiber reorientation in the response to hypertension. © 2025 The Authors.

关键词： endothelial dysfunction microstructure of extracellular matrix passive and active vascular biomechanics vascular remodeling

来源：评论

学校读者我要写书评

暂无评论

Intrusion Detection in NSL-KDD Dataset Using Hybrid Self-Organizing Map Model

引用

Computer Modeling in engineering & Sciences 2025年第4期143卷 639-671页

作者： Noveela Iftikhar Mujeeb Ur Rehman Mumtaz Ali Shah Mohammed J.F.Alenazi Jehad Ali Knowledge Unit of Systems and Technology University of Management and TechnologySialkot51310Pakistan Department of Computer Science University ofWahWah Cantt47010Pakistan Department of Computer Engineering College of Computer and Information Sciences(CCIS)King Saud UniversityRiyadh11451Saudi Arabia Department of AI Convergence Network Ajou UniversitySuwon16499Republic of Korea

Intrusion attempts against Internet of Things(IoT)devices have significantly increased in the last few *** devices are now easy targets for hackers because of their built-in security *** a Self-Organizing Map(SOM)hybrid anomaly detection system for dimensionality reduction with the inherited nature of clustering and Extreme Gradient Boosting(XGBoost)for multi-class classification can improve network traffic intrusion *** proposed model is evaluated on the NSL-KDD *** hybrid approach outperforms the baseline line models,Multilayer perceptron model,and SOM-KNN(k-nearest neighbors)model in precision,recall,and F1-score,highlighting the proposed approach’s scalability,potential,adaptability,and real-world ***,this paper proposes a highly efficient deployment strategy for resource-constrained network *** results reveal that Precision,Recall,and F1-scores rise 10%-30% for the benign,probing,and Denial of Service(DoS)*** particular,the DoS,probe,and benign classes improved their F1-scores by 7.91%,32.62%,and 12.45%,respectively.

关键词： Intrusion detection self-organizing map Internet of Things dimensionality reduction

来源：评论

学校读者我要写书评

暂无评论

Early Time Series Anomaly Prediction with Multi-Objective Optimization

引用

IEEE Transactions on Emerging Topics in Computational Intelligence 2025年第1期9卷 972-987页

作者： Chao, Ting-En Huang, Yu Dai, Hao Yen, Gary G. Tseng, Vincent S. National Yang Ming Chiao Tung University Institute of Information Management Hsinchu300093 Taiwan Oklahoma State University Department of Electrical and Computer Engineering StillwaterOK74078 United States National Yang Ming Chiao Tung University Department of Computer Science Hsinchu300093 Taiwan National Chung Hsing University Department of Management Information Systems Taichung402 Taiwan

Anomaly prediction, aiming to predict abnormal events before occurrence, plays a key role in significantly reducing costs and minimizing potential threats to mechanical devices. Monitoring machines using fixed-length time windows faces challenges in accommodating the varying characteristics of anomaly events. The lengthy and imbalanced sequence data associated with anomaly events further complicates the resolution of these challenges. Additionally, the inherent trade-off between accurate prediction and timely alarm is a crucial concern, posing difficulties in decision-making. This study puts forward a novel framework for early anomaly prediction, named early time series Anomaly Prediction with Neighbor Over-sampling and Multi-objective Optimization (APNOMO), pronounced as 'abnormal'. The framework employs three key techniques: 1) sliding windows that divide long input sequences into segments for prediction at proper intervals, 2) over-sampling and proposed neighbor over-sampling that handle imbalanced data in a novel way, and 3) multi-objective optimization that searches optimal thresholds to balance accurately prediction and timely alarm the abnormal. Experiments on a real-world dataset demonstrate APNOMO's superior performance over some state-of-the-art designs, with higher recall, F1 score, and more suitable earliness. It can predict anomalies 0.78-3.56 hours in advance, showcasing excellent early anomaly prediction capabilities for enabling predictive maintenance. © 2017 IEEE.

关键词： Predictive maintenance

来源：评论

学校读者我要写书评

暂无评论

Exploring blockchain solutions: Integrating cryptocurrency with IoT for enhanced security

Exploring blockchain solutions: Integrating cryptocurrency w...

引用

2025 International Conference on Pervasive Computational Technologies, ICPCT 2025

作者： Shravan, M. Kakad, Shital Mulani, Umar M Sharma, Geeti Tiwari, Mohit Mary Shiba, C. Sr University School of Business Telangana Warangal India Vishwakarma Institute of Technology Artificial Intelligence and Data Science Maharashtra Pune India Kj College of Engineering and Management Research Pune India India Bharati Vidyapeeth's College of Engineering Department of Computer Science and Engineering Delhi Delhi India R.M.K Engineering College Department of Computer Science and Business Systems Thiruvallur India

ISBN: (纸本)9798331508685

Blockchain, initially developed as the underlying technology for Bitcoin, has garnered significant attention for its applications beyond cryptocurrency, particularly in complex non-monetary domains. Utilizing cryptographic methods like hashing and asymmetric encryption, combined with a distributed consensus model, Blockchain-based ledgers achieve high levels of security and immutability, effectively removing the need for third-party intermediaries. Meanwhile, the rapid growth in Internet of Things (IoT) devices being connected to networks raises substantial concerns around security and privacy. Addressing these security challenges is essential for the expanding IoT landscape. This paper explores the blockchain solution by integrating cryptocurrency with IoT for enhanced security, reviewing recent studies and applications to evaluate Blockchain's role in IoT security, pinpoint challenges, and propose Blockchain-driven solutions to improve security in the IoT ecosystem. © 2025 IEEE.

关键词： Blockchain Cryptocurrency Internet of Things privacy Security

来源：评论

学校读者我要写书评

暂无评论

Phasor-FSTM:a new paradigm for multicolor super-resolution imaging of living cells based on fluorescence modulation and lifetime multiplexing

引用

Light(Science & Applications) 2025年第1期14卷 258-270页

作者： Luwei Wang Yue Chen Jiaqing Guo Xiaoyu Weng Wei Yan Jun Song Tong Ye Junle Qu Center for Biomedical Optics and Photonics&College of Physics and Optoelectronic Engineering Key Laboratory of Optoelectronic Devices and Systems of Ministry of Education and Guangdong ProvinceShenzhen UniversityShenzhen518060China The Photonics Center of Shenzhen University Shenzhen UniversityShenzhen518060China Department of Bioengineering CU-MUSC Bioengineering ProgramClemson UniversityCharlestonSouth Carolina29634USA

Multicolor microscopy and super-resolution optical microscopy are two widely used techniques that greatly enhance the ability to distinguish and resolve structures in cellular *** methods have individually transformed cellular imaging by allowing detailed visualization of cellular and subcellular structures,as well as organelle ***,integrating multicolor and super-resolution microscopy into a single method remains challenging due to issues like spectral overlap,crosstalk,photobleaching,phototoxicity,and technical *** challenges arise from the conflicting requirements of using different fluorophores for multicolor labeling and fluorophores with specific properties for super-resolution *** propose a novel multicolor super-resolution imaging method called phasor-based fluorescence spatiotemporal modulation(Phasor-FSTM).This method uses time-resolved detection to acquire spatiotemporal data from encoded photons,employs phasor analysis to simultaneously separate multiple components,and applies fluorescence modulation to create super-resolution ***-FSTM enables the identification of multiple structural components with greater spatial accuracy on an enhanced laser scanning confocal microscope using a single-wavelength *** demonstrate the capabilities of Phasor-FSTM,we performed two-color to four-color super-resolution imaging at a resolution of~λ/5 and observed the interactions of organelles in live cells during continuous imaging for a duration of over 20 *** method stands out for its simplicity and adaptability,seamlessly fitting into existing laser scanning microscopes without requiring multiple laser lines for excitation,which also provides a new avenue for other super-resolution imaging technologies based on different principles to build multi-color imaging systems with the requirement of a lower budget.

关键词： resolution lifetime simplicity

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：