检索结果-内蒙古大学图书馆

4th International Conference on Sentiment Analysis and Deep learning, ICSADL 2025

作者： Deshmukh, Santoshi Pacharaney, Utkarsha Datta Meghe Institute of Higher Education and Research Department of Artificial Intelligence and Data Science Wardha India Datta Meghe Institute of Higher Education and Research Department of Artificial Intelligence and Machine Learning Wardha India

ISBN: (纸本)9798331523923

Transparent reporting of patient outcomes is becoming more and more common as the inequalities in health outcomes amongst groups are being more widely acknowledged. In addition, documenting health outcomes over time across subgroups will be necessary for pay-for-equity programs like those put forth by the Centers for Medicare and Medicaid. One way to visualize data in the context of healthcare is through dashboards, which may be used to identify important differences in clinical outcomes, direct focused efforts toward quality improvement, and eventually enhance health equity. We assess every study that has documented the effective creation of a disparity dashboard in this viewpoint, and we also publish the information gathered and any unexpected repercussions that have been mentioned. We offer a framework for the systematic enhancement of equity via reward systems to lessen health inequalities and incentives for gathering and reporting health data. essential elements of efficient hospital management. A sophisticated approach is necessary to address the impact that hospital leadership has on patient outcomes. Because they embody patient-centered care in their everyday lives, these executives serve as the company's role models. The beliefs, vision, and leadership style of healthcare professionals greatly influence their behavior and performance. Competent leaders who inspire and empower their staff foster a culture of responsibility, creativity, and cooperation. They exhort healthcare professionals to put their patients' safety, satisfaction, and well-being first. Strategic planning, resource allocation, and decision-making are all components of good leadership that support the hospital's goal of offering the highest quality of care. essential elements of efficient hospital management. A sophisticated approach is necessary to address the impact that hospital leadership has on patient outcomes. Because they embody patient-centered care in their everyday lives, these ex

关键词： Resource allocation

来源：评论

学校读者我要写书评

暂无评论

A Comparative Analysis of Rule-based and AI-Driven Systems for Improving Customer Satisfaction and Engagement in E-Commerce using Chatbots Powered by Artificial Intelligence 6

A Comparative Analysis of Rule-based and AI-Driven Systems f...

引用

6th International Conference on Mobile Computing and Sustainable Informatics, ICMCSI 2025

作者： Deshmukh, Santoshi Gundewar, Swapnil Department of Artificial Intelligence and Data Science Datta Meghe Institute of Higher Education and Research Wardha India Department of Artificial Intelligence and Machine Learning Datta Meghe Institute of Higher Education and Research Wardha India

ISBN: (纸本)9798331522667

Chatbots: Chatbots, as robots, are the most recent and intelligent live chat communication devices. Furthermore, these days, it is commonly one of the customer service methods as it is also perfect for providing an array o f services, like personalized offers, individual communication, responsive customer support, etc. The interaction between banks and other financial institutions and their consumers has been entirely transformed due to the marriage of financial technology and artificial intelligence (AI). This scientific investigation deals with artificial intelligence (AI) technology for better communication among people;its special focus is on the effective use of chatbots and natural language processing (NLP) in banking support. Banks are increasingly turning their efforts to AI-driven chatbots as valuable means of enhancing customer support while channeling communication effectively. We are now at the point where human creativity and technology will combine and reimagine finance principles. © 2025 IEEE.

关键词： Customer satisfaction

来源：评论

学校读者我要写书评

暂无评论

On skip connections and normalisation layers in deep optimisation 23

On skip connections and normalisation layers in deep optimis...

引用

Proceedings of the 37th International Conference on Neural Information Processing Systems

作者： Lachlan E. MacDonald Jack Valmadre Hemanth Saratchandran Simon Lucey Mathematical Institute for Data Science Johns Hopkins University Australian Institute for Machine Learning University of Adelaide

We introduce a general theoretical framework, designed for the study of gradient optimisation of deep neural networks, that encompasses ubiquitous architecture choices including batch normalisation, weight normalisation and skip connections. Our framework determines the curvature and regularity properties of multilayer loss landscapes in terms of their constituent layers, thereby elucidating the roles played by normalisation layers and skip connections in globalising these properties. We then demonstrate the utility of this framework in two respects. First, we give the only proof of which we are aware that a class of deep neural networks can be trained using gradient descent to global optima even when such optima only exist at infinity, as is the case for the cross-entropy cost. Second, we identify a novel causal mechanism by which skip connections accelerate training, which we verify predictively with ResNets on MNIST, CIFAR10, CIFAR100 and ImageNet.

关键词：

来源：评论

学校读者我要写书评

暂无评论

Processes and measurements: a framework for understanding neural oscillations in field potentials

引用

Trends in Cognitive sciences 2025年第5期29卷 448-466页

作者： van Bree, Sander Levenstein, Daniel Krause, Matthew R. Voytek, Bradley Gao, Richard Department of Medicine Justus Liebig University Giessen Germany Max Planck Institute for Human Cognitive and Brain Sciences Leipzig Germany Centre for Cognitive Neuroimaging School of Psychology and Neuroscience University of Glasgow Glasgow United Kingdom MILA – Quebec AI Institute MontrealQC Canada Montreal Neurological Institute and Hospital McGill University MontrealQC Canada Department of Cognitive Science Halıcıŏglu Data Science Institute Kavli Institute for Brain & Mind University of California San Diego La JollaCA United States Machine Learning in Science Excellence Cluster Machine Learning and Tübingen AI Center University of Tübingen Tübingen Germany

Various neuroscientific theories maintain that brain oscillations are important for neuronal computation, but opposing views claim that these macroscale dynamics are ‘exhaust fumes’ of more relevant processes. Here, we approach the question of whether oscillations are functional or epiphenomenal by distinguishing between measurements and processes, and by reviewing whether causal or inferentially useful links exist between field potentials, electric fields, and neurobiological events. We introduce a vocabulary for the role of brain signals and their underlying processes, demarcating oscillations as a distinct entity where both processes and measurements can exhibit periodicity. Leveraging this distinction, we suggest that electric fields, oscillating or not, are causally and computationally relevant, and that field potential signals can carry information even without causality. © 2024 The Authors

关键词： Neurons

来源：评论

学校读者我要写书评

暂无评论

machine learning and Deep learning Approaches for Automated Diabetic Retinopathy Diagnosis

Machine Learning and Deep Learning Approaches for Automated ...

引用

2024 International Conference on Intelligent Systems for Cybersecurity, ISCS 2024

作者： Fulkar, Bhushan Burle, Rushikesh Patil, Pawan Gaurkhade, Sanskruti Gundewar, Swapnil Pacharaney, Utkarsha Datta Meghe Institute of Higher Education and Research Department of Artificial Intelligence and Data Science Wardha India Department of Artificial Intelligence and Machine Learning Maharashtra Wardha India Datta Meghe Institute of Higher Education and Research Department of Computer Science and Design Wardha India Datta Meghe Institute of Higher Education and Research Department of Artificial Intelligence and Machine Learning Wardha India

ISBN: (纸本)9798350375237

Since diabetic retinopathy (DR) is the primary cause of blindness in working-age adults worldwide, vision loss must be avoided by receiving early detection and treatment. In recent years, deep learning techniques have become more and more useful tools for automated drug discovery, possibly offering solutions to overcome the limitations of traditional screening methods. In-depth descriptions of the most recent deep learning methods for DR detection are provided in this review article. We begin by discussing the importance of DR screening, its prevalence, and the challenges faced by healthcare systems in managing this condition. Next, we discuss different deep learning architectures used for DR detection, such as CNNs and RNNs, and their combinations. We investigate how to improve the precision, resilience, and interpretability of DR detection models through the use of explainable AI, ensemble methods, transfer learning, and attention mechanisms. We also examine publicly available datasets, evaluation metrics, and preprocessing techniques that are commonly used to benchmark DR detection algorithms. We discuss the limitations and possible uses of deep learning-based deep readout detection, emphasizing the significance of large-scale, diverse datasets, the interpretability of the model, and real-world deployment concerns. The overall objectives of this review are to advance the field of automated DR detection by highlighting recent developments, compiling existing knowledge, and identifying areas that require further research. © 2024 IEEE.

关键词： Convolutional neural networks

来源：评论

学校读者我要写书评

暂无评论

Social Distancing Detection AI System using YOLO

Social Distancing Detection AI System using YOLO

引用

2023 International Conference on Computer, Electronics and Electrical Engineering and their Applications, IC2E3 2023

作者： Karthiga, M. Kavitha, D. Coimbatore Institute of Technology Dept of Artificial Intelligence and Data Science Coimbatore India Coimbatore Institute of Technology Dept. of Artificial Intelligence and Machine Learning Coimbatore India

ISBN: (纸本)9798350338003

The number of reported COVID-19 cases has once again increased due to the recent spread of the H3N2 virus. Even with vaccinations and medications, social distancing remains the most effective way to prevent the spread of these viruses. To address this issue, this paper proposes an automated system for monitoring social distancing in public places. The system utilizes a YOLOv3 object detection model to detect pedestrians in video frames captured by a camera. The frames are then converted to a top-down perspective to measure the distances between individuals in a 2D plane. If any individuals are too close to each other, the system displays a warning by highlighting them with a red frame and a red line. © 2023 IEEE.

关键词： COVID-19

来源：评论

学校读者我要写书评

暂无评论

Enhancing the Spatial Resolution of Aerial Images using a Deep Generative Approach with Multi-Stage Residual Refinement

Enhancing the Spatial Resolution of Aerial Images using a De...

引用

2024 International Conference on Advances in Modern Age Technologies for Health and Engineering science, AMATHE 2024

作者： Shree Hari, B. Joy, Emmanuel Aroma, Jenice Division Of Data Science And Cyber Security Karunya Institute Of Technology And Sciences Coimbatore India Division Of Artificial Intelligence And Machine Learning Karunya Institute Of Technology And Sciences Coimbatore India Mgk Logistics Madurai India

ISBN: (纸本)9798350371567

Aerial images are photos taken from above, usually from planes or drones, displaying a landscape as seen from above. High-quality aerial images are crucial in various tasks, like precision agriculture, post-disaster monitoring, and land use classification. However, acquiring these images can be challenging due to the constraints of underlying image sensors and several other factors. This paper introduces a deep generative approach to enhance the spatial resolution of aerial imagery. The proposed method uses an Enhanced Deep Super-Resolution (EDSR) model architecture with a multi-stage residual refinement strategy. This particular model is comprised of residual blocks with two convolutional layers and residual connections. Through a succession of these residual blocks, the model processes low-resolution, 3-channel aerial images and finally upscales the images with a convolutional layer and a depth-to-space transformation. By increasing the spatial resolution of aerial imagery, this deep generative method seeks to overcome the problems that low-quality images present in real-world applications. Findings from the experiment on the VEDAI dataset demonstrate the effectiveness of the proposed method in producing high-quality aerial images. The improvements of PSNR and SSIM metrics by 1.05% and 0.34% respectively and the reduction in MAE and NRMSE by 0.0045 and 1.23% highlight its potential impact on various fields that depend on accurate and comprehensive aerial information. The outcomes of this work will contribute to the future development of picture super-resolution techniques for use in remote sensing and geospatial analysis. © 2024 IEEE.

关键词： Antennas

来源：评论

学校读者我要写书评

暂无评论

Next-Gen Dermatology: Revolutionizing Dermatological Diagnostics with AI and Emergency Care Solutions 3

Next-Gen Dermatology: Revolutionizing Dermatological Diagnos...

引用

3rd International Conference on Intelligent and Innovative Technologies in Computing, Electrical and Electronics, IITCEE 2025

作者： Yashas, D. Kashyap, M. Shivani Charan, P.R. Hemanth Kumar, G.P. Abhishek, B.K. Deekshith, C. New Horizon College of Engineering Dept of Artificial Intelligence and Machine Learning Bengaluru India New Horizon College of Engineering Dept of Information Science and Engineering Bengaluru India Bnm Institute of Technology Dept of Artificial Intelligence and Machine Learning Bengaluru India Siddaganga Institute of Technology Dept of Artificial Intelligence and Data Science Tumkur India Vidyavardhaka College of Engineering Dept of Computer Science and Engineering Mysore India

ISBN: (纸本)9798331515911

Skin health conditions should be diagnosed and treated on time so as to prevent complications and promote better health results. This paper presents an AI-based system that makes use of Convolutional Neural Networks with the InceptionV3 Architecture for predicting skin diseases at higher levels. The model, which is based on the large DermaNet dataset, classifies skin disorders fast and effectively. The system is coupled with a simple Telegram based application for the users to get a complete solution. A user can upload a photo of the affected skin area and get a diagnosis within seconds, an explanation of the disease and how severe it is in percentage. The bot also has a location and contact sharing feature that assists users in finding nearby doctors and clinics that can provide further assistance. It further provides on the spot appropriate cures for the particular condition diagnosed. Moreover, this makes it possible for the customer to fill the healthcare void in particular the neglected ones by making it easy for them to access real time high quality services. By equipping users with relevant information and appropriate actions required at the right time, this project has showcased how AI has the ability to transform the dermatological field. © 2025 IEEE.

关键词： Convolutional neural networks

来源：评论

学校读者我要写书评

暂无评论

Expected Probabilistic Hierarchies 38

Expected Probabilistic Hierarchies

引用

38th Conference on Neural Information Processing Systems, NeurIPS 2024

作者： Kollovieh, Marcel Charpentier, Bertrand Zügner, Daniel Günnemann, Stephan School of Computation Information and Technology Technical University of Munich Germany Munich Data Science Institute Germany Munich Center for Machine Learning Germany Pruna AI Germany Microsoft Research AI for Science United States

Hierarchical clustering has usually been addressed by discrete optimization using heuristics or continuous optimization of relaxed scores for hierarchies. In this work, we propose to optimize expected scores under a probabilistic model over hierarchies. (1) We show theoretically that the global optimal values of the expected Dasgupta cost and Tree-Sampling divergence (TSD), two unsupervised metrics for hierarchical clustering, are equal to the optimal values of their discrete counterparts contrary to some relaxed scores. (2) We propose Expected Probabilistic Hierarchies (EPH), a probabilistic model to learn hierarchies in data by optimizing expected scores. EPH uses differentiable hierarchy sampling enabling end-to-end gradient descent based optimization, and an unbiased subgraph sampling approach to scale to large datasets. (3) We evaluate EPH on synthetic and real-world datasets including vector and graph datasets. EPH outperforms all other approaches quantitatively and provides meaningful hierarchies in qualitative evaluations. © 2024 Neural information processing systems foundation. All rights reserved.

关键词：

来源：评论

学校读者我要写书评

暂无评论

Adaptive Batch Sizes for Active learning: A Probabilistic Numerics Approach 27

Adaptive Batch Sizes for Active Learning: A Probabilistic Nu...

引用

27th International Conference on Artificial Intelligence and Statistics, AISTATS 2024

作者： Adachi, Masaki Hayakawa, Satoshi Jørgensen, Martin Wan, Xingchen Nguyen, Vu Oberhauser, Harald Osborne, Michael A. Machine Learning Research Group University of Oxford United Kingdom Mathematical Institute University of Oxford United Kingdom Toyota Motor Corporation Japan Department of Computer Science University of Helsinki Finland Amazon

Active learning parallelization is widely used, but typically relies on fixing the batch size throughout experimentation. This fixed approach is inefficient because of a dynamic trade-off between cost and speed—larger batches are more costly, smaller batches lead to slower wall-clock run-times—and the tradeoff may change over the run (larger batches are often preferable earlier). To address this trade-off, we propose a novel Probabilistic Numerics framework that adaptively changes batch sizes. By framing batch selection as a quadrature task, our integration-error-aware algorithm facilitates the automatic tuning of batch sizes to meet predefined quadrature precision objectives, akin to how typical optimizers terminate based on convergence thresholds. This approach obviates the necessity for exhaustive searches across all potential batch sizes. We also extend this to scenarios with constrained active learning and constrained optimization, interpreting constraint violations as reductions in the precision requirement, to subsequently adapt batch construction. Through extensive experiments, we demonstrate that our approach significantly enhances learning efficiency and flexibility in diverse Bayesian batch active learning and Bayesian optimization applications. Copyright 2024 by the author(s).

关键词： Constrained optimization

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：