检索结果-内蒙古大学图书馆

Optimal reliability–redundancy allocation of solar water pump power plant for farm

学校读者我要写书评

暂无评论

International Journal of System Assurance engineering and Management 2025年 1-9页

作者： Choudhary, Shivani Goyal, Nupur Ram, Mangey Bhandari, Ashok Singh Saini, Seema Department of Mathematics Graphic Era Deemed to Be University Dehradun India Deaprtment of Managment Greater Noida Group of Institutions MBA Institute Greater Noida India Department of Mathematics Computer Science and Engineering Graphic Era Deemed to Be University Dehradun India Department of Mathematics School of Basic & Applied Sciences Shri Guru Ram Rai University Dehradun India

The Reliability Redundancy Allocation Problem (RRAP) is a popular and important problem popular problem in reliability optimization. RRAP maximizes system reliability by assessing the redundancy and reliability variables in every subsystem with three constraints, namely, volume, cost, and weight. In this research, a solar water pump power plant for farms is developed in a series manner for RRAP. The reliability of the overall system is obtained by utilizing three metaheuristic algorithms i.e., Grew Wolf Optimization (GWO), Cuckoo Search (CS), and Hybrid of Grew Wolf Optimization and Cuckoo Search (HGWOCS). To evaluate the performance of a proposed system, the authors have compared the obtained results of each algorithm using tabular and graphical representations. The robustness of the hybrid GWO-CS is validated by the comparison among there algorithms where HGWOCS has achieved the maximum reliability 0.9723 for the proposed system.

关键词：

From RAG to QA-RAG: Integrating Generative AI for Pharmaceutical Regulatory Compliance Process 25

学校读者我要写书评

暂无评论

From RAG to QA-RAG: Integrating Generative AI for Pharmaceut...

40th Annual ACM Symposium on applied Computing, SAC 2025

作者： Kim, Jaewoong Hur, Minseok Min, Moohong Department of Applied Data Science Sungkyunkwan University Seoul Korea Republic of Department of Immersive Media Engineering/Convergence Program for Social Innovation Sungkyunkwan University Seoul Korea Republic of Department of Computer Education/Convergence Program for Social Innovation Sungkyunkwan University Seoul Korea Republic of

ISBN: (纸本)9798400706295

Regulatory compliance in the pharmaceutical industry involves navigating complex and voluminous guidelines, often requiring significant amounts of human resources. Recent advancements in Large Language Models (LLMs) and Retrieval-Augmented Generation (RAG) methods provide promising enhancements to data processing and knowledge management, potentially easing these burdens. However, despite these advancements, conventional Retrieval-Augmented Generation (RAG) methods fall short in this domain due to inherent structural problems. To address these challenges, we introduce the Question and Answer Retrieval Augmented Generation (QA-RAG) framework. This framework enhances the conventional RAG framework. It integrates a dual-track retrieval mechanism tailored to the specific and dynamic nature of pharmaceutical regulations. It utilizes not only the original query but also the answers generated by a fine-tuned LLM, thus providing a more robust foundation for document retrieval. Our experiments demonstrate that QA-RAG outperforms conventional methods in various evaluation metrics including precision, recall, and F1-score. These results underscore QA-RAG's capability to enhance both the accuracy and efficiency of regulatory compliance processes in the pharmaceutical industry. This paper details the structure and efficacy of QA-RAG, emphasizing its potential to revolutionize the regulatory compliance process in the pharmaceutical industry and beyond. Copyright © 2025 held by the owner/author(s).

关键词： Data integration

Synthesizing realistic images for improved early detection of novel pandemics: A GAN enabled approach

学校读者我要写书评

暂无评论

Heliyon 2025年第6期11卷

作者： Rounaq, Sumera Shah, Shahid Munir Aljawarneh, Mahmoud Mohammad Khan, Sarah Muhammad, Ghulam Department of Computer Science DHA Suffa University Karachi Pakistan Faculty of Engineering Sciences and Technology Hamdard University Karachi Pakistan Faculty of Information Technology Applied Science Private University Amman Jordan MEU Research Unit Middle East University Amman Jordan System Architect Low Code Solutions London United Kingdom Department of Computer Science Bahria University Karachi Pakistan

Infectious diseases like the novel Coronavirus (COVID-19) affect millions of individuals if not managed well in time. Thus, to reduce the transmission rate, effective diagnostic techniques must be identified. Early detection and diagnosis may be aided by Artificial Intelligence (AI) based automated diagnostics;however, these algorithms need a lot of pertinent data to be trained. In the early stages of a pandemic, gathering a lot of data is never an easy task. This presents challenges for accurate automatic diagnosis of such infectious diseases. To address this issue, a customized Generative Adversarial Networks (GAN) based architecture i. e., CoroWGAN-GP is proposed to synthesize images and to expand datasets. A dataset of 588 images with COVID-19, Normal, and Viral Pneumonia categories (taken from Cohen's repository [1]) was expanded using the proposed CoroWGAN-GP approach and used to train a proposed customized deep CNN architecture i. e., DeepCNN-CovNet19. Experimental results demonstrate that the proposed model achieved a considerable detection accuracy. i. e., 98.0 % in detecting the disease. The effectiveness of the proposed method was further evaluated by comparing its results with the state-of-the-art methods in the literature. Comparative analyses indicate that the proposed approach offers a promising solution for the accurate diagnosis of COVID-19-like pandemics. © 2025 The Authors

关键词： Chest X-ray Convolutional neural networks COVID-19 CXR GAN Pandemic Wasserstein GAN

Non-Destructive Ethanol-Aqueous Solution with Modified Jerusalem Resonator for Disinfectants Against COVID-19 Characterization

学校读者我要写书评

暂无评论

Non-Destructive Ethanol-Aqueous Solution with Modified Jerus...

International Symposium on Antennas and Propagation (ISAP)

作者： Patchadaporn Sangpet Prayoot Akkaraekthalin Tanaporn Pechrkool Sukanya Chudpooti Nonchanutt Chudpooti Department of Electrical and Computer Engineering Faculty of Engineering King Mongkut’s University of Technology North Bangkok Bangkok Thailand Department of Industrial Physics and Medical Instrumentation Faculty of Applied Science King Mongkut’s University of Technology North Bangkok Bangkok Thailand

This paper describes a liquid-mixture sensor based on a mushroom-shaped zeroth-order resonator (ZOR). The Jerusalem-shaped mushroom-like structure is designed to reduce the dimensions of a resonator and provide a high Q-factor with an operating frequency of 5.90 GHz. Using perturbation theory, every dielectric property of the liquid-mixture influences the resonance frequency. The structure of the sensor is composed of the modified Jerusalem resonator and 3D-printed case for locking the liquid under test (LUT) to reduce the problematic error from position measurement. Five Ethanol-aqueous concentrations, e.g., 60%, 65%, 70%, 75% and 80%, are chosen to investigate the responsibility of the modified Jerusalem resonator. By detecting the change in resonant frequency of the reflection coefficient, S11, the measured results can classify five conditions of the LUT. Other benefits of the zeroth-order resonator sensor include a nondestructive method, real-time monitoring, and no life-cycle limitation.

关键词：

Gesture-Aware Zero-Shot Speech Recognition for Patients with Language Disorders

学校读者我要写书评

暂无评论

arXiv 2025年

作者： Kim, Seungbae Lee, Daeun Stark, Brielle Han, Jinyoung Department of Computer Science and Engineering University of South Florida United States Yale School of Medicine Yale University United States Department of Speech Language and Hearing Sciences Indiana University Bloomington United States Department of Applied Artificial Intelligence Sungkyunkwan University Korea Republic of

Individuals with language disorders often face significant communication challenges due to their limited language processing and comprehension abilities, which also affect their interactions with voice-assisted systems that mostly rely on Automatic Speech Recognition (ASR). Despite advancements in ASR that address disfluencies, there has been little attention on integrating non-verbal communication methods, such as gestures, which individuals with language disorders substantially rely on to supplement their communication. Recognizing the need to interpret the latent meanings of visual information not captured by speech alone, we propose a gesture-aware ASR system utilizing a multimodal large language model with zero-shot learning for individuals with speech impairments. Our experiment results and analyses show that including gesture information significantly enhances semantic understanding. This study can help develop effective communication technologies, specifically designed to meet the unique needs of individuals with language impairments. © 2025, CC BY-NC-ND.

关键词： Zero-shot learning

Machine Learning For An Explainable Cost Prediction of Medical Insurance

学校读者我要写书评

暂无评论

arXiv 2023年

作者： Orji, Ugochukwu Ukwandu, Elochukwu Department of Computer Science Faculty of Physical Science University of Nigeria Enugu State Nsukka Nigeria Department of Applied Computing and Engineering Cardiff School of Technologies Cardiff Metropolitan University Wales CardiffCF5 2YB United Kingdom

Predictive modeling in healthcare continues to be an active actuarial research topic as more insurance companies aim to maximize the potential of Machine Learning (ML) approaches to increase their productivity and efficiency. In this paper, the authors deployed three regression-based ensemble ML models that combine variations of decision trees through Extreme Gradient Boosting (XGBoost), Gradient-boosting Machine (GBM), and Random Forest (RF) methods in predicting medical insurance costs. Explainable Artificial Intelligence (XAi) methods SHapley Additive exPlanations (SHAP) and Individual Conditional Expectation (ICE) plots were deployed to discover and explain the key determinant factors that influence medical insurance premium prices in the dataset. The dataset used comprised 986 records and is publicly available in the KAGGLE repository. The models were evaluated using four performance evaluation metrics, including R-squared (R2), Mean Absolute Error (MAE), Root Mean Squared Error (RMSE), and Mean Absolute Percentage Error (MAPE). The results show that all models produced impressive outcomes;however, the XGBoost model achieved a better overall performance although it also expanded more computational resources, while the RF model recorded a lesser prediction error and consumed far fewer computing resources than the XGBoost model. Furthermore, we compared the outcome of both XAi methods in identifying the key determinant features that influenced the PremiumPrices for each model and whereas both XAi methods produced similar outcomes, we found that the ICE plots showed in more detail the interactions between each variable than the SHAP analysis which seemed to be more high-level. It is the aim of the authors that the contributions of this study will help policymakers, insurers, and potential medical insurance buyers in their decision-making process for selecting the right policies that meet their specific needs. © 2023, CC BY-NC-SA.

关键词： Machine learning

Generalization of CNNs on Relational Reasoning with Bar Charts

学校读者我要写书评

暂无评论

arXiv 2025年

作者： Cui, Zhenxing Chen, Lu Wang, Yunhai Haehn, Daniel Wang, Yong Pfister, Hanspeter School of Computer Science and Technology Shandong University China State Key Lab of CAD&CG Zhejiang University China School of Information Renmin University of China China College of Science and Mathematics University of Massachusetts Boston United States College of Computing and Data Science Nanyang Technological University Singapore School of Engineering and Applied Sciences Harvard University United States

This paper presents a systematic study of the generalization of convolutional neural networks (CNNs) and humans on relational reasoning tasks with bar charts. We first revisit previous experiments on graphical perception and update the benchmark performance of CNNs. We then test the generalization performance of CNNs on a classic relational reasoning task: estimating bar length ratios in a bar chart, by progressively perturbing the standard visualizations. We further conduct a user study to compare the performance of CNNs and humans. Our results show that CNNs outperform humans only when the training and test data have the same visual encodings. Otherwise, they may perform worse. We also find that CNNs are sensitive to perturbations in various visual encodings, regardless of their relevance to the target bars. Yet, humans are mainly influenced by bar lengths. Our study suggests that robust relational reasoning with visualizations is challenging for CNNs. Improving CNNs’ generalization performance may require training them to better recognize task-related visual properties. Copyright © 2025, The Authors. All rights reserved.

关键词： Convolutional neural networks

Deep Reinforcement Learning of Robotic Manipulation for Whip Targeting

学校读者我要写书评

暂无评论

Deep Reinforcement Learning of Robotic Manipulation for Whip...

IEEE International Conference on Simulation, Modeling, and Programming for Autonomous Robots (SIMPAR)

作者： Xiang Bai Junyi Wang Xiaofeng Xiong Evangelos Boukas Applied Mathematics and Computer Science Technical University of Denmark Copenhagen Denmark Automation and Control Group Electrical and Photonics Engineering Technical University of Denmark Copenhagen Denmark SDU Biorobotics Mærsk Mc-Kinney Møller Institute University of Southern Denmark Odense Denmark

ISBN: (数字)9798331516857

ISBN: (纸本)9798331516864

This paper aims at a deep reinforcement learning (DRL) controller for fast (< 1.5s) manipulation of a flexible tool (i.e., whip) to hit a target in 3D space. The controller consists of a DRL algorithm for optimizing joint motions, and a proportional-derivative (PD) mechanism for tracking the optimized motions. Their objective is to minimize the distance between the whip-end-tip and the target. The proposed controller was validated in a 7-DOF robot arm by comparing four DRL algorithms in the physical simulator MuJoCo. It shows that the proximal policy optimization (PPO) outperforms others by obtaining the maximum average reward. Notably, PPO can still effectively interact with the environment under sparse or even unrewarding conditions, making it a robust choice for complex and dynamic tasks. Our work provides preliminary knowledge of DRL applications to fast robotic arm control in flexible object manipulation.

关键词： Three-dimensional displays Target tracking Heuristic algorithms Programming Aerospace electronics Deep reinforcement learning Manipulators Optimization Autonomous robots

Towards Fleet-wide Sharing of Wind Turbine Condition Information through Privacy-preserving Federated Learning

学校读者我要写书评

暂无评论

arXiv 2022年

作者： Jenkel, L. Jonas, S. Meyer, A. Bern University of Applied Sciences School of Engineering and Computer Science Biel Switzerland Università della Svizzera Italiana Faculty of Informatics Viganello Lugano Switzerland

Terabytes of data are collected by wind turbine manufacturers from their fleets every day. And yet, a lack of data access and sharing impedes exploiting the full potential of the data. We present a distributed machine learning approach that preserves the data privacy by leaving the data on the wind turbines while still enabling fleet-wide learning on those local data. We show that through federated fleet-wide learning, turbines with little or no representative training data can benefit from more accurate normal behavior models. Customizing the global federated model to individual turbines yields the highest fault detection accuracy in cases where the monitored target variable is distributed heterogeneously across the fleet. We demonstrate this for bearing temperatures, a target variable whose normal behavior can vary widely depending on the turbine. We show that no turbine experiences a loss in model performance from participating in the federated learning process, resulting in superior performance of the federated learning strategy in our case studies. The distributed learning increases the normal behavior model training times by about a factor of ten due to increased communication overhead and slower model convergence. © 2022, CC BY-NC-ND.

关键词： Wind turbines