检索结果-内蒙古大学图书馆

Electronics and Renewable Systems (ICEARS), International Conference on

作者： Shaik Mohammad Waseem Jeshwanth N L Lingesh A Matta Jayanth Bhargavi C H Batini Dhanwanth Dept. of Computer Science and Engineering Panimalar Institute of Technology (Anna University) Chennai India Dept. of Artificial Intelligence and Data Science Panimalar Institute of Technology (Anna University) Chennai India Dept. of Computer Science and Engineering Sai Spurthi Institute of Technology India Dept. of AI Engineering Jönköping University Jönköping Sweden

ISBN: (数字)9798331509675

ISBN: (纸本)9798331509682

This study explores the use of Genetic Algorithms (GA) to solve the NP-hard combinatorial optimization problem known as the Travelling Salesman Problem (TSP). The suggested GA, called GA-P, performs better than conventional techniques because it uses a new randomized crossover threshold. According to experimental assessments, GA-P outperforms a normal Genetic Algorithm with fixed crossover (GA-N, 628.6) and approaches the best Branch and Bound solutions (B&B, 560.6) with an average cost of 617.4 for 16-node graphs. With average execution times of 0.1347 seconds for 16 nodes, GA-P also demonstrates constant computational efficiency, outperforming B&B's exponentially growing temporal complexity (169.37 seconds for 16 nodes). In order to balance solution quality and computational cost, hyperparameter tweaking revealed the ideal values of population percentage (pp = 1.1), crossover proportion (cp = 0.8), mutation threshold (mt = 0.3), and exploration probability (ep = 0.2). The results show that GA-P is a scalable and efficient substitute for TSP, with room for improvement via parallelization and exploratory techniques.

关键词： Renewable energy sources Costs Evolutionary computation Computational efficiency Complexity theory Optimization Tuning Genetic algorithms

来源：评论

学校读者我要写书评

暂无评论

BN-Pool: a Bayesian Nonparametric Approach to Graph Pooling

arXiv

引用

arXiv 2025年

作者： Castellana, Daniele Bianchi, Filippo Maria Dept. of Statistics Computer Science and Applications Università degli Studi di Firenze Italy Dept. of Mathematics and Statistics UiT the Arctic University of Norway Norway NORCE Norwegian Research Centre AS Norway

We introduce BN-Pool, the first clustering-based pooling method for Graph Neural Networks (GNNs) that adaptively determines the number of supernodes in a coarsened graph. By leveraging a Bayesian non-parametric framework, BN-Pool employs a generative model capable of partitioning graph nodes into an unbounded number of clusters. During training, we learn the node-to-cluster assignments by combining the supervised loss of the downstream task with an unsupervised auxiliary term, which encourages the reconstruction of the original graph topology while penalizing unnecessary proliferation of clusters. This adaptive strategy allows BN-Pool to automatically discover an optimal coarsening level, offering enhanced flexibility and removing the need to specify sensitive pooling ratios. We show that BN-Pool achieves superior performance across diverse benchmarks. Copyright © 2025, The Authors. All rights reserved.

关键词： Graph neural networks

来源：评论

学校读者我要写书评

暂无评论

Role of IoT in Smart Cities: A Review, applications, Open Challenges and Solutions

Role of IoT in Smart Cities: A Review, Applications, Open Ch...

引用

Electronics and Renewable Systems (ICEARS), International Conference on

作者： Kolla Vivek Gurrala Swetha V T Ram Pavan Kumar M Anusuri Krishna Veni Neti Praveen Masthan Rao Kale Dept. of CSE QIS College of Engineering & Technology (A) Ongole Andhra Pradesh India Dept. of CSE Malla Reddy College of Engineering and Technology (A) Telangana India Dept. of Computer Science Kakaraparti Bhavanarayana College Vijayawada Andhra Pradesh India Dept. of Computer Science & Engineering – Data Science Madanapalle Institute of Technology & Science Madanapalle Andhra Pradesh India Dept. of Computer Science and Design SRKR Engineering College Bhimavaram Andhra Pradesh India Dept. of CSE Pace Institute of Technology and Sciences (A) Ongole Andhra Pradesh India

ISBN: (数字)9798331509675

ISBN: (纸本)9798331509682

Smart cities have emerged as a response to the fast expansion of metropolitan areas; these places strive to improve the lives of their residents while simultaneously promoting environmentally responsible growth. To maximize urban services, the Internet of Things (IoT) is an essential technology that allows for the seamless integration of sensors, devices, and systems. This allows smart cities to become a reality. With an emphasis on smart transportation, energy management, healthcare, environmental monitoring, and public safety, this paper explores the critical role of IoT in smart cities. data security, privacy, scalability, and interoperability are some of the problems that arise when smart cities use IoT, despite the revolutionary potential it has. These obstacles are thoroughly examined in this work, along with possible solutions to these problems. This study gives a thorough overview of the IoT and its impact on smart city development, innovation, and urban resilience by focusing on the present situation, applications, and unanswered questions.

关键词： Waste management data privacy Technological innovation Smart cities Scalability data security Medical services Public security Internet of Things Interoperability

来源：评论

学校读者我要写书评

暂无评论

Predicting Lung Cancer with Artificial Neural Network: Analyzing Performance and Feature Importance

Predicting Lung Cancer with Artificial Neural Network: Analy...

引用

Cognitive Computing in Engineering, Communications, sciences and Biomedical Health Informatics (IC3ECSBHI), International Conference on

作者： Vatsal Mehta Varun M Ashwini Kodipalli Roshan Joy Martis Shridhar B. Devamane Ushasree. A Dept. of AI & Data Science Global Academy of Technology Bangalore Karnataka India Dept. of Electronics & Communication Engineering Manipal Institute of Technology Bengaluru Manipal Academy of Higher Education Manipal India Department of Computer Science School of Computing Mysuru Dept. of Computer Science & Engineering Gokaraju Lailavathi Engineering College Hyderabad India

ISBN: (数字)9798331518523

ISBN: (纸本)9798331518530

Lung cancer is the primary cause of cancer-related deaths globally. This cancer is majorly observed in smokers, increasing air pollution has also begun to affect non-smokers. Research indicates that exposure to polluted air significantly raises the risk of developing lung cancer, posing a major threat to public health. Despite advancements in medical technology, the mortality rate from lung cancer remains high, highlighting the importance of early diagnosis. Machine learning and deep learning offer promising solutions for early and accurate prediction through advanced data analysis. In our study, we utilized a dataset with 26 attributes for discern lung cancer, categorizing cases into three risk levels: high, medium, and low. We applied various algorithms and develop an Artificial Neural Network (ANN), comparing the overall performance with various machine-learning and deep-learning models. Our results show that the ANN achieved a high accuracy of 94.9% using the Adam optimizer, making it an effective model for lung cancer prediction.

关键词： Deep learning Accuracy Computational modeling Biological system modeling Lung cancer Mortality Artificial neural networks Predictive models Prediction algorithms Public healthcare

来源：评论

学校读者我要写书评

暂无评论

Simulating Mobile Robot Vision: An Analysis of RGB-D Versus RGB-Based Distance Accuracy and CPU Optimization

Simulating Mobile Robot Vision: An Analysis of RGB-D Versus ...

引用

International Conference on Artificial Intelligence in Information and Communication (ICAIIC)

作者： Minseok Kong Jiho Park Daye Lee Nikolaos Kourtzanidis Jungmin So Dept. Computer Science and Engineering Sogang University Seoul South Korea Dept. Artificial Intelligence Dongguk University Seoul South Korea Dept. Data Science Seoul National University Seoul South Korea Cyberworks Robotics Toronto Canada

ISBN: (数字)9798331507022

ISBN: (纸本)9798331507039

From the perspective of mobile robot vision, to facilitate secure and seamless interactions between the robot and its environment, it is paramount to attain precise and metic-ulous distance estimation between the robot and the detected entities. Moreover, in the absence of GPUs, mobile robots are compelled to rely solely on CPU resources to manage vision-related tasks and other operational imperatives. To address the challenges above, we focused on reducing CPU usage in the vision system and analyzing distance estimation accuracy in two scenarios: leveraging RGB-D and RGB cameras. We executed real-time object detection targeting humans utilizing YOLOv8 Nano, subsequently undertaking a comparative analysis between the distance estimates derived through monocular dept. estimation using a MobileNetV2 model, trained on the KITTI dataset, and the distance measurements obtained from RGB-D cameras. Additionally, we optimized the deep learning models to reduce CPU usage, employing a hardware-specific technique, OpenVINO conversion, and both a hardware-specific and a hardware-agnostic strategy, Post-Training Quantization (PTQ). Consequently, we reduced CPU utilization by approximately 4.62% in the RGB-D scenario and 5.32% in the RGB scenario.

关键词： Accuracy Machine vision dept. measurement Robot vision systems Estimation Object detection Cameras Real-time systems Mobile robots Optimization

来源：评论

学校读者我要写书评

暂无评论

Supply Chain 4.0: Driving Operational Efficiency, Resilience, and Cybersecurity Using Advanced Technologies

Supply Chain 4.0: Driving Operational Efficiency, Resilience...

引用

Computational Intelligence Approaches and applications (ICCIAA), International Conference on

作者： Bilal Ibrahim Sowan Saad Zighan Ahmad Altarawneh Dept. of Bus. Intell. & Data Analytics University of Petra Amman Jordan Dept. of Management and Marketing Qatar University Doha Qatar Dept. of Computer Science Brunel University of London London UK

ISBN: (数字)9798331523657

ISBN: (纸本)9798331523664

Supply Chain Industry 4.0 denotes the transformative evolution of global supply chain systems, integrating advanced digital technologies with complex interconnected networks. It enhances operational efficiency, agility, and global reach. Nevertheless, it also presents intensive, fast-developing cyber and operational risks. Ransomware, phishing attacks, and supply chain poisoning are notable vulnerabilities, alongside Internet of Things (IoT)-related issues, such as unauthorized access to IoT devices and sensors. These challenges are further compounded by insider threats, including intentional sabotage or accidental mistakes, which degrade operational integrity, continuity, and resilience. The advancements in Supply Chain Industry 4.0 are driven by technologies such as blockchain, federated learning, Trusted Execution Environments (TEEs), and advanced Artificial Intelligence applications. These innovations enable better traceability, real-time decision-making, and improved supply chain resilience. This paper examines the complexities of Supply Chain Industry 4.0 by adopting a multi-layered strategy that analyzes how these advanced technologies strengthen security protocols, bridge interoperability gaps, and mitigate vulnerabilities. Also, by incorporating detailed case studies and rigorous technical evaluations, the study highlights the effectiveness of these solutions in ensuring robust supply chain operations. Furthermore, the research considers emerging opportunities and challenges, providing a holistic assessment of current practices and implications for future research. These findings offer actionable pathways to enhance the resilience and adaptability of global supply chain ecosystems in an increasingly volatile and interconnected environment.

关键词： Technological innovation Protocols Phishing Supply chains Real-time systems Fourth Industrial Revolution Blockchains Internet of Things Ransomware Resilience

来源：评论

学校读者我要写书评

暂无评论

Leveraging Frozen Foundation Models and Multimodal Fusion for BEV Segmentation and Occupancy Prediction

IEEE Open Journal of Vehicular Technology

引用

IEEE Open Journal of Vehicular Technology 2025年 6卷 1241-1261页

作者： Hayes, Seamie Sistu, Ganesh Eising, Ciaran University of Limerick Dept. of Electronic and Computer Engineering Castletroy Co. Limerick V94 T9PX Ireland University of Limerick SFI CRT Foundations in Data Science Castletroy Co. Limerick V94 T9PX Ireland Research Centre Castletroy Co. Limerick V94 T9PX Ireland

In Bird's Eye View perception, significant emphasis is placed on deploying well-performing, convoluted model architectures and leveraging as many sensor modalities as possible to reach maximal performance. This paper investigates whether foundation models and multi-sensor deployments are essential for enhancing BEV perception. We examine the relative importance of advanced feature extraction versus the number of sensor modalities and assess whether foundation models can address feature extraction limitations and reduce the need for extensive training data. Specifically, incorporating the self-supervised DINOv2 for feature extraction and Metric3Dv2 for dept. estimation into the Lift-Splat-Shoot framework results in a 7.4 IoU point increase in vehicle segmentation, representing a relative improvement of 22.4%, while requiring only half the training data and iterations compared to the original model. Furthermore, using Metric3Dv2's dept. maps as a pseudo-LiDAR point cloud within the Simple-BEV model improves IoU by 2.9 points, marking a 6.1% relative increase compared to the Camera-only setup. Finally, we extend the famous Gaussian Splatting BEV perception models, GaussianFormer and GaussianOcc, through multimodal deployment. The addition of LiDAR information in GaussianFormer results in a 9.4-point increase in mIoU, a 48.7% improvement over the Camera-only model, nearing state-of-the-art multimodal performance even with limited LiDAR scans. In the self-supervised GaussianOcc model, incorporating LiDAR leads to a 0.36-point increase in mIoU, representing a 3.6% improvement over the Camera-only model. This limited gain can be attributed to the absence of LiDAR encoding and the self-supervised nature of the model. Overall, our findings highlight the critical role of foundation models and multi-sensor integration in advancing BEV perception. By leveraging sophisticated foundation models and multi-sensor deployment, we can further model performance and reduce data requirem

关键词： Gaussian distribution

来源：评论

学校读者我要写书评

暂无评论

Predicting Customer Purchase Intent Using Logistic Regression: A data-Driven Approach

Predicting Customer Purchase Intent Using Logistic Regressio...

引用

Machine Learning and Autonomous Systems (ICMLAS), International Conference on

作者： C. Rajarajachozhan G. Karthika A. Rajakrishnammal M. Mano P. Nithish Kumar E. Jayahari Dept. of Electronics and Communication Engineering Saveetha Engineering College Chennai Tamil Nadu Dept. of Artificial Intelligence and Data Science Saveetha Engineering College Chennai Tamil Nadu Dept. of Computer Science and Engineering Saveetha Engineering College Chennai Tamil Nadu

ISBN: (数字)9798331505745

ISBN: (纸本)9798331505752

Predicting customer purchase intent is crucial for modern businesses in order to improve engagement, refine marketing strategies, and increase revenue. Conventional rule-based systems encounter difficulties with dynamic and extensive customer data, limiting their flexibility and precision. This project employs a machine learning methodology utilizing logistic regression to forecast customer intent based on demographic, transactional, and behavioral data. The proposed model analyses input features including age, browsing history, and previous purchases, employing sophisticated data preprocessing methods such as One-Hot Encoding and feature scaling to guarantee data integrity. Through hyperparameter tuning and the assessment of critical metrics including accuracy, precision, recall, and Fl-score, the system provides dependable and actionable insights. Real-time predictive capabilities enable businesses to customize customer interactions, facilitating targeted marketing and enhancing customer experience. This project underscores the capacity of predictive analytics to tackle real-world issues, facilitating data-driven decision-making and sustainable business development.

关键词： Logistic regression Adaptation models Accuracy Pipelines Companies Predictive models Real-time systems data models Predictive analytics Business

来源：评论

学校读者我要写书评

暂无评论

A Machine Learning and data Analytics Approach to Patient Risk Stratification

A Machine Learning and Data Analytics Approach to Patient Ri...

引用

data science, Agents & Artificial Intelligence (ICDSAAI), International Conference on

作者： Veeramalai Sankaradass V.K Manindra Manish Rajesh Anandhan Ramprakash Velmurugan Sabarinathan Sakthivel Dept. of Computer Science and Engineering Chennai Institute of Technology Chennai India Dhanalakshmi Srinivasan Institute of Medical Sciences and Hospital Perambalur India Dept. of Artificial Intelligence and Data Science Chennai Institute of Technology Chennai India

ISBN: (数字)9798331537555

ISBN: (纸本)9798331537562

In this paper, we introduce a robust machine learning (ML) and data analytic solution for risk stratification of patients to promote clinical decision support systems (CDSS) that provide enhanced protection of the patient health interests. Our approach, which we have shown would minimally intrude into peoples' lives, uses predictive models to analyze patient data to identify high-risk patients early in the history of a wide range of medical conditions. Utilizing integrated structured data (e.g., lab results, vital signs) and unstructured data (e.g., clinical notes), our system improves the accuracy of medical decisions made by healthcare providers. We use a combination of supervised learning algorithms and real-time data processing to recalibrate risk scores, which serve as actionable insights for anticipatory interventions. The strategy also comes with extensive validation protocols to verify the efficacy of her model in a healthcare environment. Our results indicate that this practice dramatically increases the accuracy of risk assessments, with important implications for patient care and resource utilization.

关键词： Decision support systems data analysis Accuracy Protocols Supervised learning Real-time systems Risk management Resource management Protection Random forests

来源：评论

学校读者我要写书评

暂无评论

Diversity Seeking Techniques for Red-Teaming Large Language Models

Diversity Seeking Techniques for Red-Teaming Large Language ...

引用

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

作者： Seokhan Lee Bonhwa Ku Hanseok Ko Dept. Computer Science and Engineering Korea University Seoul Republic of Korea Machine Learning Big Data Institute Korea University Seoul Republic of Korea Dept. Electrical and Computer Engineering Korea University Seoul Republic of Korea

ISBN: (数字)9798350368741

ISBN: (纸本)9798350368758

In this paper, we present new techniques for increasing the diversity of red-teaming prompts generated by automated machine learning-based methods, thereby enabling the discovery of more vulnerabilities in large language models. Using reinforcement learning to train models to output effective prompts for this task results in the models converging deterministically to a single output. Our first technique, which we term Defender, acts by blocking the reward signal for prompts that have already been discovered, thus making what was a stationary problem into a non-stationary problem that compels the reward maximizing algorithm to continually seek new prompts. Our second technique, Teamplay, trains two prompt generation models in tandem and adds the KL divergence between them to the reward in order to make them search in disparate regions of the space of prompts. Our techniques are shown experimentally to increase the effectiveness and diversity of prompts generated by existing reinforcement learning baselines.

关键词： Learning systems Large language models Diversity reception Signal processing algorithms Reinforcement learning Signal processing Acoustics Safety Speech processing

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：