检索结果-内蒙古大学图书馆

6th International Conference on Big data and Artificial Intelligence, BDAI 2023

作者： Bokolo, Biodoumoye George Chen, Lei Liu, Qingzhong Sam Houston State University Dept. of Computer Science HuntsvilleTX United States Georgia Southern University Dept. of Info. Tech. Statesboro Georgia

ISBN: (纸本)9798350303667

This paper presents an approach for cybercriminal profiling using pre-trained DistilBert, LSTM, and BERT models. By analyzing criminal behaviors and linking them to offender characteristics, the proposed method utilizes structural and parameter learning techniques. Digital forensics, as a means to locate criminal and cybercriminal activity, is highlighted as increasingly important. The suggested strategy incorporates tools such as technical competency tests, a dynamic criminal knowledge base, and visualization to provide investigators with a comprehensive understanding of the case. The paper also discusses the potential benefits of integrating this approach into a cloud-based infrastructure, offering a faster and more cost-effective solution. © 2023 IEEE.

关键词： Cybersecurity

来源：评论

学校读者我要写书评

暂无评论

Genetic data for Early Detection and Treatment of Cystic Fibrosis through Machine Learning

Genetic Data for Early Detection and Treatment of Cystic Fib...

引用

International Conference on I-SMAC (IoT in Social, Mobile, Analytics and Cloud) (I-SMAC)

作者： Deepthi P Bhuvaneshwari M Chithra K Dhanasree Lp Jananee V J.A. Adlin Layola Dept. of Computer Science and Engineering Rajalakshmi Engineering College Chennai India

ISBN: (数字)9798350376425

ISBN: (纸本)9798350376432

Cystic Fibrosis (CF) is a genetic disorder that significantly impacts the respiratory and digestive systems, requiring early detection and precise management to improve patient outcomes. This research study presents a machine learning model designed to predict the presence, severity, and potential treatments for Cystic Fibrosis using genetic data. By leveraging collaborative filtering techniques, our model not only identifies individuals at risk but also enables personalized treatment plans based on genetic predispositions. The proposed model aims to revolutionize Cystic Fibrosis diagnosis and management by automating the screening process on a mass scale. This automation facilitates the identification of demographic areas and genetic populations with a higher susceptibility to Cystic Fibrosis, enabling targeted interventions. Through large-scale genetic data analysis, our approach ensures that at-risk populations receive timely and appropriate care, ultimately improving the quality of life for Cystic Fibrosis patients. The findings suggest that the implementation of this predictive model in clinical settings can lead to earlier and accurate diagnoses, personalized treatment plans, and significant advancements in Cystic Fibrosis care. The implications of this research extend beyond Cystic Fibrosis, offering a framework for applying machine learning in the diagnosis and treatment of other genetic disorders.

关键词： Adaptation models Machine learning algorithms Accuracy Biological system modeling Predictive models Genetics Prediction algorithms data models Real-time systems Random forests

来源：评论

学校读者我要写书评

暂无评论

Evaluating Neural Word Embeddings for Sanskrit 18

Evaluating Neural Word Embeddings for Sanskrit

引用

18th World Sanskrit Conference

作者： Sandhan, Jivnesh Paranjay, Om Adideva Komal, Digumarthi Behera, Laxmidhar Goyal, Pawan Dept. of Electrical Engineering IIT Kanpur India Tata Consultancy Services India Dept. of Electrical and Systems Engineering University of Pennsylvania United States Dept. of Computer Science and Engineering IIT Kharagpur India

ISBN: (纸本)9781950737710

Recently, the supervised learning paradigm's surprisingly remarkable performance has garnered considerable attention from Sanskrit Computational Linguists. As a result, the Sanskrit community has put laudable efforts to build task-specific labeled data for various downstream Natural Language Processing (NLP) tasks. The primary component of these approaches comes from representations of word embeddings. Word embedding helps to transfer knowledge learned from readily available unlabelled data for improving task-specific performance in low-resource setting. Last decade, there has been much excitement in the field of digitization of Sanskrit. To effectively use such readily available resources, it is essential to perform a systematic study on word embedding approaches for the Sanskrit language. In this work, we investigate the effectiveness of word embeddings. We classify word embeddings in broad categories to facilitate systematic experimentation and evaluate them on four intrinsic tasks. © 2022 Proceedings of the Computational Sanskrit and Digital Humanities at the 18th World Sanskrit Conference. All rights reserved.

关键词： Embeddings

来源：评论

学校读者我要写书评

暂无评论

Accelerating Real-Time Imaging for Radiotherapy: Leveraging Multi-GPU Training with PyTorch 22

Accelerating Real-Time Imaging for Radiotherapy: Leveraging ...

引用

22nd IEEE International Conference on Machine Learning and Applications, ICMLA 2023

作者： Obe, Ruth Kaufmann, Brandt Baird, Kaelen Kadel, Sam Soltani, Yasmin Cham, Mostafa Gobbert, Matthias K. Barajas, Carlos A. Jiang, Zhuoran Sharma, Vijay R. Ren, Lei Peterson, Stephen W. Poif, Jerimy C. U. of Houston-Clear Lake Dept. of Computer Science and of Software Engineering United States U. of San Francisco Dept. of Mathematics and Statistics United States Skidmore College Dept. of Computer Science and of Mathematics United States Mount Holyoke College Dept. of Computer Science and of Psychology United States U. of Houston Dept. of Biomedical Engineering United States U. of Maryland Dept. of Information Systems Baltimore County United States U. of Maryland Dept. of Mathematics and Statistics Baltimore County United States Medical Physics Graduate Program Duke University United States U. of Maryland School of Medicine Dept. of Radiation Oncology United States U. of Maryland School of Medicine Department of Radiation Oncology United States U. of Cape Town Dept. of Physics South Africa H3D Inc. United States

ISBN: (纸本)9798350345346

Proton beam therapy is an advanced form of cancer radiotherapy that uses high-energy proton beams to deliver precise and targeted radiation to tumors. This helps to mit-igate unnecessary radiation exposure in healthy tissues. Real-time imaging of prompt gamma rays with Compton cameras has been suggested to improve therapy efficacy. However, the camera's non-zero time resolution leads to incorrect interaction classifications and noisy images that are insufficient for accurately assessing proton delivery in patients. To address the challenges posed by the Compton camera's image quality, machine learning techniques are employed to classify and refine the generated data. These machine-learning techniques include recurrent and feedforward neural networks. A PyTorch model was designed to improve the data captured by the Compton camera. This decision was driven by PyTorch's flexibility, powerful capabilities in handling sequential data, and enhanced G PU usage. This accelerates the model's computations on large-scale radiotherapy data. Through hyperparameter tuning, the validation accuracy of our PyTorch model has been improved from an initial 7 % to over 60 %. Moreover, the PyTorch Distributed data Parallelism strategy was used to train the RNN models on multiple G PU s, which significantly reduced the training time with a minor impact on model accuracy. © 2023 IEEE.

关键词： Clas-sification Compton camera Distributed data Parallelism Proton beam therapy PyTorch Recurrent neural network

来源：评论

学校读者我要写书评

暂无评论

An Adaptive Clustering Scheme for Client Selections in Communication-efficient Federated Learning

arXiv

引用

arXiv 2025年

作者： Chen, Yan-Ann Chen, Guan-Lin Dept. of Computer Science and Engineering Yuan Ze University Taoyuan Taiwan

Federated learning is a novel decentralized learning architecture. During the training process, the client and server must continuously upload and receive model parameters, which consumes a lot of network transmission resources. Some methods use clustering to find more representative customers, select only a part of them for training, and at the same time ensure the accuracy of training. However, in federated learning, it is not trivial to know what the number of clusters can bring the best training result. Therefore, we propose to dynamically adjust the number of clusters to find the most ideal grouping results. It may reduce the number of users participating in the training to achieve the effect of reducing communication costs without affecting the model performance. We verify its experimental results on the non-IID handwritten digit recognition dataset and reduce the cost of communication and transmission by almost 50% compared with traditional federated learning without affecting the accuracy of the model. © 2025, CC BY-NC-ND.

关键词： Hierarchical clustering

来源：评论

学校读者我要写书评

暂无评论

A Novel Integrated IoT Framework with Classification Approach for Medical data Analysis 9

A Novel Integrated IoT Framework with Classification Approac...

引用

9th International Conference on Computing for Sustainable Global Development, INDIACom 2022

作者： Bikku, Thulasi Sree, K. P. N. V. Satya Jarugula, Jyothi Sunkara, Mounika Vignan's Nirula Institute of Technology and Science for Women Dept. of Computer Science and Engineering Guntur India Usha Rama College of Engineering and Technology Dept. of Computer Science and Engineering Vijayawada India Usha Rama College of Engineering and Technology Department of Cse Vijayawada India

ISBN: (纸本)9789380544441

In the healthcare systems, usage of advanced integrated technologies like Internet of things (IoT) and Machine learning (ML) techniques were limited. Different amalgams of IoT devices and ML mechanisms are available for medical sector but are limited to certain domains only. These models either provide patients current state data or specific domain analyzing and surveillance pre/post treatment data like heart or brain functions with corresponding medical aid. Also data available is used as clinical study for medical professionals and for better understandings of patients about their state. Specific domain gadgets' like wrist bands or smart bands uses some sensors about vital, temperature and pulse etc., checkups are available but they were not meant for diagnosing or for treatments. In this paper, we proposed an integrated model to use IoT and ML algorithms for a healthcare system. Tracking of patients' status can be done using some sensors such as lightweight, portable, and low-powered sensor nodes. These Sensors sense the patient's status and send the parametric data to the central controller, to take actions during the critical condition of the patients. The data sent to the controller always provided in secure and encryption form. At the same time, patient data is sent to doctors, so that they can provide the instructions to the caretakers of the patients with quick and proper solutions in real-time. For disease prediction, our model uses supervised machine learning algorithms, In order to get the efficient feature set and improve the better accuracy, and pre-processing techniques to eliminate features that are irrelevant, missing values and outliers from biomedical data which aids in better disease prediction. To further strengthen the proposed integrated model design is compared with various traditional classification algorithms to specify its improved accuracy and computational time for accurate prediction of the patient's disease and acts as a decision suppo

关键词： Internet of things

来源：评论

学校读者我要写书评

暂无评论

Solving Task-Select Problems Using a Quantum Annealer

Solving Task-Select Problems Using a Quantum Annealer

引用

IEEE International Conference on Consumer Electronics (ICCE)

作者： Koki Mita Yuta Yachi Keisuke Fukada Nozomu Togawa Dept. of Computer Science and Communications Engineering Waseda University Tokyo Japan

ISBN: (数字)9798331521165

ISBN: (纸本)9798331521172

To tackle environmental and social challenges, we face numerous tasks that need to be solved. However, due to limited time and resources, it is unrealistic to solve all tasks. Therefore, prioritizing tasks to ensure efficient handling and meeting criteria is essential. When the number of tasks is large, prioritizing the optimal combination of tasks on classical computers becomes highly time-consuming. We define this pri-oritization as a combinatorial optimization problem, termed the task-select problem, and address it using a quantum annealer. A quantum annealer is specialized hardware designed to efficiently solve combinatorial optimization problems by transforming them into Quadratic Unconstrained Binary Optimization (QUBO) problems. In this paper, we propose a QUBO formulation for the task-select problem. Experimental evaluations demonstrate that our approach enables solving the task-select problem with up to 100 items using a real quantum annealer.

关键词： computers Annealing Benchmark testing Quantum annealing Hardware Optimization Faces Consumer electronics

来源：评论

学校读者我要写书评

暂无评论

Innovative Approaches to CT Image Enhancement: Deep Learning Solutions for Denoising and Deblurring

Innovative Approaches to CT Image Enhancement: Deep Learning...

引用

Ambient Intelligence in Health Care (ICAIHC), International Conference on

作者： Gopal Krishan Prajapat Dharmendra Yadav Dept. of Computer Science and Engineering Bikaner Technical University Bikaner India

ISBN: (数字)9798331530259

ISBN: (纸本)9798331530266

Computed Tomography (CT) is essential in medical diagnostics, providing detailed images of internal structures. Noise and blur introduced during image acquisition can com-promise diagnostic accuracy, especially in low-dose CT (LDCT) scans aimed at reducing radiation exposure. Deep learning methods have emerged as effective solutions for addressing these challenges, with significant advancements in both denoising and deblurring techniques. This paper reviews recent deep learning approaches for CT image enhancement, covering supervised and self-supervised methods. The paper also explores emerging trends such as diffusion models and transformer-based architectures that offer improved image quality and restoration efficiency. Simultaneous denoising and deblurring techniques are examined for their potential in clinical applications. Promising future directions include the use of large-scale pretrained models, zero-shot learning, and prompt-based frameworks, which have the potential to further enhance CT image quality while maintaining low radiation exposure, thereby advancing diagnostic capabilities.

关键词： Image quality Adaptation models Accuracy Computed tomography Computational modeling Noise reduction Noise Zero shot learning Convolutional neural networks Image enhancement

来源：评论

学校读者我要写书评

暂无评论

3D CNN Implementation in Kinesis Recognition & Sign Language

3D CNN Implementation in Kinesis Recognition & Sign Language

引用

2024 OPJU International Technology Conference on Smart Computing for Innovation and Advancement in Industry 4.0, OTCON 2024

作者： Chaudhary, Mayank Mahajan, Aryan Raj, Gaurav Sar, Ayan Choudhury, Tanupriya Pant, Bhasker Dewangan, Bhupesh Kumar Sharda University Dept. of Computer Science Greater Noida201310 India School of Computer Science Uttarakhand Dehradun248007 India Graphic Era Deemed to Be University Uttarakhand Dehradun248002 India Graphic Era Deemed to Be University Cse Dept Uttarakhand Dehradun248002 India Op Jindal University Department of Computer Science and Engineering Chhattisgarh Raigarh India

ISBN: (纸本)9798350373783

The 3D Convolutional Neural Network (CNN) is used for Kinesis Recognition in this paper. The suggested model exhibits enhanced recognition accuracy of intricate hand and body movements by acquiring spatiotemporal information from kinesis sequences. The network can more reliably recognize complex motions because it can simultaneously utilize spatial and temporal information. By means of comprehensive testing and assessment, this work demonstrates the efficacy of the 3D CNN architecture in real-time kinesis categorization, indicating that it is a viable option for a variety of uses, including virtual reality, sign language interpretation, and human-computer interaction. © 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

Mapping Rice Paddy Areas in Nueva Ecija Using Temporal Sentinel -1 SAR Imagery

Mapping Rice Paddy Areas in Nueva Ecija Using Temporal Senti...

引用

2024 Symposium on Insight to Foresight via Geospatial Technologies

作者： Cueto, Jann Rovic Allard, Jessah Mei Obras, Johanne Mae De Asis, Allysa Maxine Galut, Arnel Dayrit, Roland Tandoc, John Allen Purio, Mark Angelo Dept. of Earth and Space Science Rizal Technological University Boni Mandaluyong Philippines Dept. of Geology Caraga State University Butuan Philippines Dept. of Electronics Engineering Adamson University Manila Philippines Space Technologies and Applications Research Laboratory Adamson University Manila Philippines

Remote Sensing is proven to be helpful in various ways, and for an agricultural country like the Philippines, mapping in farmlands is not that common. Using the Sentinel 1 Synthetic Aperture Radar (SAR) data, the study reveals rice paddy classification brought by supervised machine learning on the municipality of Guimba, in Nueva Ecija - the Rice Bowl Capital of the country. The study in Guimba, Nueva Ecija, addresses the absence of comprehensive rice classification studies, focusing on creating a classified rice paddy map and assessing spectral indices (NDMI, NDVI, NDWI) using synthetic aperture radar and optical imager. Results show successful differentiation between rice and non-rice paddies. Temporal analysis emphasizes monitoring water availability, soil moisture, and vegetation health. Despite signs of overprediction in the CART model, its effectiveness in mapping rice paddies is notable. The spatial distribution maps contribute to targeted monitoring, enabling efficient interventions and improved agricultural practices. This research highlights the inherent use of remote sensing in rice crop management, offering valuable insights for farmers and country's food security. © Author(s) 2024.

关键词： Soil moisture

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：