检索结果-内蒙古大学图书馆

IEEE/ACM 21st international conference on Mining Software Repositories (MSR)

作者： Jiang, Wenxin Yasmin, Jerin Jones, Jason Synovic, Nicholas Kuo, Jiashen Bielanski, Nathaniel Tian, Yuan Thiruvathukal, George K. Davis, James C. Purdue Univ W Lafayette IN 47907 USA Queens Univ Kingston ON Canada Loyola Univ Chicago Chicago IL USA

ISBN: (纸本)9798350363982;9798400705878

The development and training of deep learning models have become increasingly costly and complex. Consequently, software engineers are adopting pre-trained models (PTMs) for their downstream applications. The dynamics of the PTM supply chain remain largely unexplored, signaling a clear need for structured datasets that document not only the metadata but also the subsequent applications of these models. Without such data, the MSR community cannot comprehensively understand the impact of PTM adoption and reuse. This paper presents the PeaTMOSS dataset, which comprises metadata for 281,638 PTMs and detailed snapshots for all PTMs with over 50 monthly downloads (14,296 PTMs), along with 28,575 open-source software repositories from GitHub that utilize these models. Additionally, the dataset includes 44,337 mappings from 15,129 downstream GitHub repositories to the 2,530 PTMs they use. To enhance the dataset's comprehensiveness, we developed prompts for a large language model to automatically extract model metadata, including the model's training datasets, parameters, and evaluation metrics. Our analysis of this dataset provides the first summary statistics for the PTM supply chain, showing the trend of PTM development and common shortcomings of PTM package documentation. Our example application reveals inconsistencies in software licenses across PTMs and their dependent projects. PeaTMOSS lays the foundation for future research, offering rich opportunities to investigate the PTM supply chain. We outline mining opportunities on PTMs, their downstream usage, and crosscutting questions. Our artifact is available at https://***/PurdueDualityLab/PeaTMOSS-Artifact. Our dataset is available at https://***/file-manager?origin_id=ff978999-16c2-4b50-ac7a-947ffdc3eb1d& origin_path=%2F.

关键词： datasets Machine learning Deep neural networks Model zoos Package registries Open-source Empirical software engineering

来源：评论

学校读者我要写书评

暂无评论

Machine learning Approaches for Improving the Survival Prediction Rate in Heart Failure Patients 21

Machine Learning Approaches for Improving the Survival Predi...

引用

21st OITS international conference on Information Technology, OCIT 2023

作者： Singh, Shrejal Rout, Minakhi Lana, Rohan Das, Suchismita Jena, Junali Jasmine Kumar Acharya, Anuja Bihar Engineering University Siwan College of Engg Management Bihar India KIIT Deemed to Be University School of Computer Engineering Bhubaneswar India

ISBN: (纸本)9798350358230

This research work aims to highlight the significance of dimensionality reduction and minority class oversampling techniques for improving the performance metrics result of a machine learning problem involving the survival prediction rate in heart failure patients. Methods: The authors have implemented various existing feature selection and feature extraction methods to find the right subset of features for enhancing the model performance. To tackle the problem of class imbalance present in the data, three different minority oversampling methods are used. A Random Forest model is trained on each new subset of features with a 10-fold cross-validated AUC-ROC score considered as the performance evaluation metric. Results: In our experiments, feature selection methods performed better than feature extraction methods for increasing the area under the ROC curve. Recursive Feature Elimination produced the best results among all the experimented dimensionality reduction techniques. 'Borderline-SMOTE and Borderline-SMOTE SVM' oversampling techniques performed better than ADASYN in almost all combinations. © 2023 IEEE.

关键词： Feature Selection

来源：评论

学校读者我要写书评

暂无评论

Toward Knowledge engineering Using MatFlow for Inverse Quantum Dye Design 25

Toward Knowledge Engineering Using MatFlow for Inverse Quant...

引用

25th IEEE international conferences on High Performance Computing and Communications, 9th international conference on data Science and Systems, 21st IEEE international conference on Smart City and 9th IEEE international conference on Dependability in Sensor, Cloud and Big data Systems and Applications, HPCC/DSS/SmartCity/DependSys 2023

作者： Rahman, Hasan H. Jamil, Hasan M. University of Idaho Department of Computer Science United States

ISBN: (纸本)9798350330014

Inverse materials design can be viewed as a reverse engineering process of material constituents from a user given set of targets characterizing material properties. Machine learning techniques could be used to map a feature space x of material properties to a contextual latent space z over an object space O, and then reconstruct a distinct object space O′ over the latent space z to maximize the context c. In this paper, we introduce a new machine learning tool, called MatFlow, for inverse material design using machine learning. In MatFlow, we first learn the latent space z characterizing a context feature set c. MatFlow then helps identify novel materials O′ over the latent space z with a potential to exceed the contextual threshold θ without destabilizing the latent space. We explain MatFlow features and its capabilities using an application in quantum dye material discovery. The main focus of this paper is developing a computational strategy to identify z in the context of c, and inform the characteristics of z space values of O′ most likely to meet the contextual threshold θ. © 2023 IEEE.

关键词： Knowledge representation

来源：评论

学校读者我要写书评

暂无评论

Tactical Training Using Augmented Reality/Virtual Reality and Haptics 4

Tactical Training Using Augmented Reality/Virtual Reality an...

引用

4th international conference on Technological Advancements in Computational Sciences, ICTACS 2024

作者： Vashisht, Sanchit Chitkara University Institute of Engineering and Technology Chitkara University Department of Computer Science and Engineering Punjab India

ISBN: (纸本)9798350387490

This study investigates the effects of these immersive technologies on military personnel's training efficacy, efficiency, and safety. Virtual reality (VR) immerses users in artificial surroundings, augmented reality (AR) superimposes digital data on the actual world, and haptics adds tactile input to improve realism. These technologies work together to create training simulations that are incredibly realistic, adaptable, and closely mimic real-world situations. Examples of these scenarios include military operations, emergency response protocols, and technical maintenance jobs. This research emphasizes the advantages of AR, VR, and haptics in military training, including higher information retention, improved skill learning, and enhanced preparation for operational difficulties. It does this by doing a thorough assessment of the available literature and case studies. By eliminating the need for live exercises and giving students practical experience in authentic settings, these immersive technologies provide a secure and affordable substitute for conventional training approaches. Furthermore, employees stationed in remote or inaccessible regions may receive training nearly anytime, anywhere, thanks to the scalability and mobility of AR and VR devices. In the future, this field of study will focus on developing increasingly complex simulation environments and on utilizing advances in machine learning and artificial intelligence to build intelligent and adaptable training systems. Furthermore, haptics, AR, and VR are being widely used by allied agencies and military groups, which has great potential for collaborative training platforms and more equitable access to excellent training materials. Incorporating immersive technology into tactical and technical training signifies a fundamental change in military readiness and education, with far-reaching effects on the 21st-century armed forces' efficacy and adaptability. © 2024 IEEE.

关键词： Personnel training

来源：评论

学校读者我要写书评

暂无评论

MedNetV3: A Deep learning Model for MRI Brain Tumor Detection with MobileNetV3 and Advanced data Augmentation 21

MedNetV3: A Deep Learning Model for MRI Brain Tumor Detectio...

引用

21st international Computer conference on Wavelet Active Media Technology and Information Processing, ICCWAMTIP 2024

作者： Tariq, Hussain Ahmed Fengli, Zhang Salim Shakib, Muhamud Md Javed, Iram School of Information and Software Engineering University of Electronic Science and Technology of China China University of Agriculture Department of Computer Science Faisalabad Pakistan

ISBN: (纸本)9798331519254

In this paper, we present a novel deep learning model medical network (MedNetV3) developed for brain tumor detection. It incorporates advanced data augmentation techniques based on the MobileNetV3 architecture. MedNetV3 is developed for higher classification accuracy at the cost of slightly lower computational efficiency and is therefore suitable for clinical use. The proposed model overcomes the class imbalance problem and can make the model robust to the variations of MRI scans included in the extensive data augmentation methods, such as random rotations, flips. The proposed model trained and evaluated using a publicly available MRI dataset. It achieved a high Area Under the Curve (AUC) value of 0.99 for both the Precision-Recall and Receiver Operating Characteristic (ROC) curves. The results show that MedNetV3 is effective in correctly distinguishing brain tumors and therefore has the potential to be used in medical image classification. © 2024 IEEE.

关键词： Deep learning

来源：评论

学校读者我要写书评

暂无评论

Enhancing Sustainable Mobility: Lane Change strategies for Efficient Traffic Flow at stationary Bottlenecks

Enhancing Sustainable Mobility: Lane Change Strategies for E...

引用

2024 ASU international conference in Emerging Technologies for Sustainability and intelligent Systems, ICETSIS 2024

作者： Badshah, Imran Saad, Syed Ammad, Syed Rasheed, Kumeel Khan, Muhammad Waqas Ali, Zawar University of Eng. & Technology Dept. of Urban Infrastructure Plan. Peshawar Pakistan Cecos University Dept. of Civil Engineering Peshawar Pakistan Griffith University Dept. of Civil Engineering Gold Coast Australia

ISBN: (纸本)9798350372229

The 21st century witnesses a surge in global population, leading to increased road traffic. This study focuses on understanding factors impacting time delays at road bottlenecks. data from Peshawar, Pakistan, regarding density, speed, and delay patterns in peak traffic were collected. A lane change model examines delay-causing parameters, including density, speed, queue length, and pedestrians. Utilizing Camlytics software, peak period entry/exit times were compared to ideal and off-peak scenarios. Higher destination route density raises delays and lowers speed. This research offers insights into road bottleneck delays. This study aimed to investigate and classify the key factors contributing to time delays at stationary bottlenecks on roads. The research collected data on density, speed, and delay patterns during morning and afternoon peak traffic periods in Peshawar city. The future is focused on embracing technology, data-driven regulations, and cooperative efforts for long-term fixes. © 2024 IEEE.

关键词： stationary Bottleneck Sustainable Traffic Flow Target Lane Traffic Models

来源：评论

学校读者我要写书评

暂无评论

Real-Time Illuminance Assessment Web Application for Energy Efficiency Systems 1

Real-Time Illuminance Assessment Web Application for Energy ...

引用

1st international conference on Innovative and intelligent Information Technologies, IC3IT 2024

作者： Abderraouf, Seniguer Mustapha, Aouache Abdelhamid, Iratni Lmse Laboratory University Mohamed El Bachir El Ibrahimi of Bordj Bou Arreridj Bordj Bou Arreridj34030 Algeria Division Télécom Baba Hassen DZ Algiers16303 Algeria University Mohamed El Bachir El Ibrahimi of Bordj Bou Arreridj Faculty of Science and Technology Department of Electrical Engineering DZ Bordj Bou Arreridj34030 Algeria

ISBN: (纸本)9798331541064

The growing emphasis on energy efficiency in building management has led to increased adoption of daylight harvesting systems, which optimize the exploitation of sunlight to minimize the energy consumption of artificial lighting. Accurate lighting assessment is crucial for the effective implementation of these systems. This paper presents a mobile-friendly web application designed to comprehensively assess lighting conditions in daylight harvesting systems. The application leverages smartphone sensors and deep learning technology to measure illuminance levels, providing real-time data and analytics to facility managers and building occupants. Key features include user-friendly interfaces for data visualization, integration with building management systems, and automated reporting functions. Preliminary field tests demonstrate the potential of the app to improve energy savings and optimize occupant comfort by facilitating the precise calibration of daylight-responsive lighting controls. The proposed solution represents a cost-effective and accessible tool for advancing sustainable building practices and supports the broader adoption of smart lighting technologies. © 2024 IEEE.

关键词： Artificial intelligence Computer Vision Deep learning Energy Efficiency Indoor Light Web Application

来源：评论

学校读者我要写书评

暂无评论

Fast and Reliable Modeling of Offshore Wind Generation for Adequacy studies

引用

IEEE TRANSACTIONS ON INDUstRY APPLICATIONS 2023年第6期59卷 7116-7125页

作者： Nguyen, Thuy-hai Toubeau, Jean-Francois De Jaeger, Emmanuel Vallee, Francois Univ Mons Elect Power Engn Unit B-7000 Mons Belgium Catholic Univ Louvain B-1348 Louvain La Neuve Belgium

Considering the increasing proportion of offshore wind generation in the energy mix, it becomes essential to properly account for aerodynamic effects that impact the power extracted from the wind. Indeed, due to computational constraints, offshore wind energy is currently modelled in a very simple and approximate way in adequacy studies, neglecting important factors such as wake effects. Hence, in this paper, data-driven proxy models are developed for learning the complex relation between free flow wind information and the resulting aggregated output power of wind farms. Those supervised Machine learning-based models are used as fast and reliable surrogates of wake models, embedding their ability to describe the wind and turbines behavior, but with much lower computational times. These models are then included in an adequacy study built upon sequential Monte-Carlo simulations. The collected results are compared with those obtained with the current simplified modelling approach for offshore generation. We observe the importance of accurately representing intra-farm aerodynamic effects since reliability indices can be significantly underestimated when using the simplified modelling, thus hiding potential stressed conditions within the power system.

关键词： Adequacy machine learning offshore wind VARMA wake effects

来源：评论

学校读者我要写书评

暂无评论

Research on the Expert database System of Characteristic Species in Wuling Mountain Area 21

Research on the Expert Database System of Characteristic Spe...

引用

21st international Computer conference on Wavelet Active Media Technology and Information Processing, ICCWAMTIP 2024

作者： Yujun, Yang Yimei, Yang Wang, Zhou Liyun, Li Ling, Peng Jin, Xiang Huixia, Shu College of Computer and Artificial Intelligence Software College Huaihua University Huaihua418008 China Hunan Provincial Key Laboratory of Ecological Agriculture Intelligent Control Technology Huaihua418008 China School of Computer and Software Engineering Xihua University Chengdu610039 China

ISBN: (纸本)9798331519254

The Wuling Mountains Area(WMA) is one of the important ecological protection areas in central China, known for its rich biodiversity and unique ecological environment. To effectively protect the species resources in this area and strengthen ecological monitoring and research, this paper proposes and designs the 'Wuling Mountains Region Characteristic Species Expert database System.' By integrating expert resources, species data, ecological environment information, and modern information technologies, an intelligent system has been developed that combines data management, expert support, species identification, and research analysis. This paper discusses in detail the system's requirements analysis, design concept, implementation technologies, and potential applications, aiming to provide technological support for biodiversity conservation and sustainable ecological development in the region. © 2024 IEEE.

关键词： Characteristic species Expert database system intelligent system Species data Wuling Mountain Area

来源：评论

学校读者我要写书评

暂无评论

Self-supervised Heterogeneous Graph Representation learning with Attribute Completion 21

Self-supervised Heterogeneous Graph Representation Learning ...

引用

21st international Computer conference on Wavelet Active Media Technology and Information Processing, ICCWAMTIP 2024

作者： Yajie, Lei School of Computer Science and Engineering University of Electronic Science and Technology of China Sichuan Province Chengdu611731 China

ISBN: (纸本)9798331519254

The missing attribute problem in heterogeneous graph representation learning has received increasing attention recently. However, previous methods fail to address the following issues. First, they pre-train the guidance information to complete the missing data before representation learning, which brings extra cost and may lead to sub-optimal completion. Second, they conduct node pair sampling for contrastive learning, which also causes extra computational cost. To tackle these issues, in this paper, we propose a new self-supervised heterogeneous graph learning method with attribute completion. Specifically, we propose a heterogeneous layer to learn the topology embeddings as the guidance. We then complete missing attributes by weighted aggregation. After that, we generate two graph views and conduct representation learning. We investigate the contrastive constraint by pushing the correlation matrix of two views close to the identity matrix. Moreover, we reconstruct the original node attributes to improve the effectiveness of representation learning. We conduct experiments to verify the superiority of our method. © 2024 IEEE.

关键词： Self-supervised learning

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：