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： Gayaki, Unnati Daronde, Subodh Tale, Abhay Barhate, Aditya Datta Meghe Institute of Higher Education and Research Faculty of Engineering and Technology Department of Computer Science and Engineering Maharashtra Wardha India Faculty of Engineering and Technology Department of Biomedical Engineering Maharashtra Wardha India Datta Meghe Institute of Higher Education and Research Faculty of Engineering and Technology Department of Artificial Intelligence and Data Science Maharashtra Wardha India Datta Meghe Institute of Higher Education and Research Faculty of Engineering and Technology Department of Artificial Intelligence and Machine Learning Maharashtra Wardha India

ISBN: (纸本)9798331523923

Diabetes is a prevalent and chronic disease affecting millions worldwide, posing significant challenges in its management and treatment. This review article aims to explore the current and potential future roles of machine learning (ML) in diabetes management, focusing on progress, applications, and challenges. It discusses various ML techniques, including support vector machines (SVM), logistic regression, decision trees, ensemble methods, and advanced deep learning approaches such as long short-term memory (LSTM) networks and convolutional neural networks (CNN). These methods have demonstrated benefits in analytics, patient monitoring, decision support systems, personalised treatment, and patient engagement. However, significant challenges persist, such as computational complexity, data quality, limited interoperability of systems, and ethical concerns regarding AI use in healthcare. The article addresses these challenges by fostering collaboration between AI implementers and health care professionals. This collaboration is essential to develop long-term, legally compliant, and solution-oriented frameworks that enhance patient care and diabetes management and pave the way for more efficient and ethical AI applications in healthcare. © 2025 IEEE.

关键词： Support vector machines

来源：评论

学校读者我要写书评

暂无评论

Tumor Synthesis Conditioned on Radiomics

Tumor Synthesis Conditioned on Radiomics

引用

2025 IEEE/CVF Winter Conference on Applications of computer Vision, WACV 2025

作者： Kim, Jonghun Na, Inye Ko, Eun Sook Park, Hyunjin Sungkyunkwan University Department of Electrical and Computer Engineering Suwon Korea Republic of Samsung Medical Center Sungkyunkwan University School of Medicine Department of Radiology Center for Imaging Science Suwon Korea Republic of

ISBN: (纸本)9798331510831

Due to privacy concerns, obtaining large datasets is challenging in medical image analysis, especially with 3D modalities like Computed Tomography (CT) and Magnetic Resonance Imaging (MRI). Existing generative models, developed to address this issue, often face limitations in output diversity and thus cannot accurately represent 3D medical images. We propose a tumor-generation model that utilizes radiomics features as generative conditions. Radiomics features are high-dimensional handcrafted semantic features that are biologically well-grounded and thus are good candidates for conditioning. Our model employs a GAN-based model to generate tumor masks and a diffusion-based approach to generate tumor texture conditioned on radiomics features. Our method allows the user to generate tumor images according to user-specified radiomics features such as size, shape, and texture at an arbitrary location. This enables the physicians to easily visualize tumor images to better understand tumors according to changing radiomics features. Our approach allows for the removal, manipulation, and repositioning of tumors, generating various tumor types in different scenarios. The model has been tested on tumors in four different organs (kidney, lung, breast, and brain) across CT and MRI. The synthesized images are shown to effectively aid in training for downstream tasks and their authenticity was also evaluated through expert evaluations. Our method has potential usage in treatment planning with diverse synthesized tumors. Our code is available at ***/jongdoryITS-Radiomics. © 2025 IEEE.

关键词： Photointerpretation

来源：评论

学校读者我要写书评

暂无评论

Detection of Neurogenic Voice Disorders Using the Fisher Vector Representation of Cepstral Features

引用

Journal of Voice 2025年第3期39卷 757-763页

作者： Yagnavajjula, Madhu Keerthana Alku, Paavo Rao, Krothapalli Sreenivasa Mitra, Pabitra Department of Signal Processing and Acoustics Aalto University Espoo Finland Advanced Technology Development Centre Indian Institute of Technology Kharagpur West Bengal Kharagpur India Department of Computer Science and Engineering Indian Institute of Technology Kharagpur West Bengal Kharagpur India

Neurogenic voice disorders (NVDs) are caused by damage or malfunction of the central or peripheral nervous system that controls vocal fold movement. In this paper, we investigate the potential of the Fisher vector (FV) encoding in automatic detection of people with NVDs. FVs are used to convert features from frame level (local descriptors) to utterance level (global descriptors). At the frame level, we extract two popular cepstral representations, namely, Mel-frequency cepstral coefficients (MFCCs) and perceptual linear prediction cepstral coefficients (PLPCCs), from acoustic voice signals. In addition, the MFCC features are also extracted from every frame of the glottal source signal computed using a glottal inverse filtering (GIF) technique. The global descriptors derived from the local descriptors are used to train a support vector machine (SVM) classifier. Experiments are conducted using voice signals from 80 healthy speakers and 80 patients with NVDs (40 with spasmodic dysphonia (SD) and 40 with recurrent laryngeal nerve palsy (RLNP)) taken from the Saarbruecken voice disorder (SVD) database. The overall results indicate that the use of the FV encoding leads to better identification of people with NVDs, compared to the defacto temporal encoding. Furthermore, the SVM trained using the combination of FVs derived from the cepstral and glottal features provides the overall best detection performance. © 2022 The Authors

关键词： Fisher vector Glottal features MFCC Neurogenic voice disorders Perceptual linear prediction Support vector machine

来源：评论

学校读者我要写书评

暂无评论

Stress Detection Using SVM and Virtual Assistant

Stress Detection Using SVM and Virtual Assistant

引用

2025 IEEE International Students' Conference on Electrical, Electronics and computer science, SCEECS 2025

作者： Amale, Pranav Vibhute, Anup Mathpati, Mahesh S. Jatti, Ashwini V Thigale, Somnath B. Dr. D. Y. Patil Institute of Technology Electronics and Telecommunications Department Pune India Dr. D. Y. Patil Institute of Technology Instrumentation Engineering Department Pune India SVERIs College of Engineering Electronics and Telecommunications Department Pandharpur India Fabtech College of Engineering Computer Science and Engineering Department Sangola India

ISBN: (纸本)9798331529833

This paper presents a chatbot model designed to detect stress in users through simple yes-or-no questions. Utilizing a Support Vector Machine (SVM) algorithm, the model achieves high accuracy in identifying stress levels and provides users with actionable advice or medical recommendations. The chatbot's user-friendly interface is built using Anaconda Navigator and Python, ensuring ease of use for non-technical users. The model achieves a 100% accuracy rate on the given dataset, making it a promising tool for stress detection and early intervention. © 2025 IEEE.

关键词： Python

来源：评论

学校读者我要写书评

暂无评论

I3S: Importance Sampling Subspace Selection for Low-Rank Optimization in LLM Pretraining

arXiv

引用

arXiv 2025年

作者： Zhang, Haochen Yin, Junze Wang, Guanchu Liu, Zirui Zhang, Tianyi Shrivastava, Anshumali Yang, Lin Braverman, Vladimir Department of Computer Science Rice University Houston United States Electrical and Computer Engineering Department University of California Los Angeles Los Angeles United States Department of Computer Science John Hopkins University Baltimore United States

Low-rank optimization has emerged as a promising approach to enabling memory-efficient training of large language models (LLMs). Existing low-rank optimization methods typically project gradients onto a low-rank subspace, reducing the memory cost of storing optimizer states. A key challenge in these methods is identifying suitable subspaces to ensure an effective optimization trajectory. Most existing approaches select the dominant subspace to preserve gradient information, as this intuitively provides the best approximation. However, we find that in practice, the dominant subspace stops changing during pretraining, thereby constraining weight updates to similar subspaces. In this paper, we propose importance sampling subspace selection (I3S) for low-rank optimization, which theoretically offers a comparable convergence rate to the dominant subspace approach. Empirically, we demonstrate that I3S significantly outperforms previous methods in LLM pretraining tasks. © 2025, CC BY-NC-SA.

关键词：

来源：评论

学校读者我要写书评

暂无评论

Application Interface for GPenSIM

Application Interface for GPenSIM

引用

1st International Symposium on Systems Modelling and Simulation, SMS 2024

作者： Feng, Yuming Samolej, Slawomir Davidrajuh, Reggie School of Computer Science and Engineering Chongqing Three Gorges University Wenzhou China Department of Computer and Control Engineering Rzeszow University of Technology Rzeszow Poland Department of Electrical Engineering and Computer Science University of Stavanger Stavanger Norway

ISBN: (纸本)9789819646128

The General-purpose Petri Net Simulator (GPenSIM) is a tool used by researchers to model, simulate, and analyze the performance of discrete event systems. Its popularity stems from its simple interface, extensibility, and compatibility with the MATLAB platform. This paper focuses on the crucial role of GPenSIM in Petri Net model-based software-to-software and software-to-hardware access and control, highlighting unique features and capabilities in its application interface. © The Author(s), under exclusive license to Springer Nature Singapore Pte Ltd. 2025.

关键词： MATLAB

来源：评论

学校读者我要写书评

暂无评论

Lightweight Adaptive Learning Algorithm for Energy-Latency Tradeoff in IoMT-Enabled Edge Computing 17

Lightweight Adaptive Learning Algorithm for Energy-Latency T...

引用

17th International Conference on Machine Vision, ICMV 2024

作者： Yadav, Rahul Shafiq, Muhammad Kumar, Mohit Wei, Li Elaziz, Mohamed Abd College of Computer Science and Technology Harbin Engineering University Harbin China Cyberspace Institute of Advance Technology Guangzhou University Guangzhou China Department of Information Technology National Institute of Technology Jalandhar India Department of Mathematics Faculty of Science Zagazig University Zagazig Egypt

ISBN: (纸本)9781510688278

Smart Healthcare is witnessing widespread implementation of Internet-of-Medical-Things (IoMT) devices. These devices play a crucial role in collecting vast amounts of data from various smart healthcare applications, which are then processed to facilitate informed decision-making. Edge computing has emerged as a valuable platform offering computational resources to handle this data collection efficiently. However, a critical challenge arises from the inappropriate and inefficient classical approaches to fair resource allocation in these energy-intensive, short battery life and delay-intolerant portable devices. To address this issue, this paper presents a Lightweight Adaptive Learning Offloading Algorithm designed to optimize energy and latency in an edge computing environment. The proposed algorithm formulates the problem as a combined minimization of latency and energy costs while satisfying constraints related to limited battery capacity and service latency deadlines. To achieve this, the algorithm employs the adaptation parameters (α, β, γ), and the tradeoff target (C′) is configured based on the specific task requirements. Experimental results demonstrate the advantages of the proposed scheme, including energy savings and minimized latency in an edge environment. To validate the effectiveness of the proposed scheme under a realistic scenario using the iFogSim2 simulator. © 2025 SPIE.

关键词： Electronic health record

来源：评论

学校读者我要写书评

暂无评论

Segmentation and Classification of Unharvested Arecanut Bunches Using Deep Learning 1st

Segmentation and Classification of Unharvested Arecanut Bunc...

引用

1st International Conference on Intelligent Systems in Computing and Communications, ISCComm 2023

作者： Dhanesha, R. Umesha, D.K. Hiremath, Gurudeva Shastri Girish, G.N. Shrinivasa Naika, C.L. Department of Studies in Computer Science Davangere University Karnataka Davangere India Shivasharane Akkamahadevi Governament Womens Polytechnic Shiralakoppa Karnataka Shivamogga India Department of Computer Science and Engineering SJEC Karnataka Mangalore India Computer Science and Engineering Group Indian Institute of Information Technology Chittor Sricity India Department of Studies in Computer Science and Engineering University B.D.T. College of Engineering Karnataka Davangere India

ISBN: (纸本)9783031756047

One of the most profitable commercial palms in India is the arecanut. The market price of arecanuts, which is determined by the arecanuts maturity level, is the only factor that determines the profitability of arecanut plantation farmers. To reduce the financial loss of the farmer, knowing maturity level of the unharvested arecanut bunch is a tedious job due to lack of expertise. We presented an automated technique to determine the maturity level of the unharvested arecanut bunches in order to address this problem. The proposed method works on two phases: first, an optimized U-Net model was used for semantic segmentation of the arecanut bunch images then the classification of the segmented arecanut bunches was performed using transfer learning approach. Experiments were conducted on RGB and saturation channel of HSV color space images, and a comparative study of RGB and saturation channel of HSV color space is presented. Result of the experiments show that the segmentation and classification of the arecanut bunches from input image is efficient using RGB color space. © The Author(s), under exclusive license to Springer Nature Switzerland AG 2025.

关键词： Semantic Segmentation

来源：评论

学校读者我要写书评

暂无评论

WritingRing: Enabling Natural Handwriting Input with a Single IMU Ring 25

WritingRing: Enabling Natural Handwriting Input with a Singl...

引用

Proceedings of the 2025 CHI Conference on Human Factors in Computing Systems

作者： Zhe He Zixuan Wang Chun Yu Chengwen Zhang Xiyuan Shen Yuanchun Shi Department of Computer Science and Technology Tsinghua University Beijing Beijing China Department of Computer Science and Technology Tsinghua University Beijing China Paul G. Allen School of Computer Science & Engineering University of Washington Seattle Washington USA and Department of Computer Science and Technology Tsinghua University Beijing China Department of Computer science and Technology Tsinghua University Beijing China and Qinghai University Xining China

来源：评论

学校读者我要写书评

暂无评论

PTMs-TSCIL: Pre-Trained Models Based Class-Incremental Learning

arXiv

引用

arXiv 2025年

作者： Wu, Yuanlong Nie, Mingxing Zhu, Tao Chen, Liming Ning, Huansheng Wan, Yaping School of Computer Science University of South China 421001 China School of Computer Science and Technology Dalian University of Technology 116024 China Department of Computer & Communication Engineering University of Science and Technology Beijing100083 China

Class-incremental learning (CIL) for time series data faces critical challenges in balancing stability against catastrophic forgetting and plasticity for new knowledge acquisition, particularly under real-world constraints where historical data access is restricted. While pre-trained models (PTMs) have shown promise in CIL for vision and NLP domains, their potential in time series class-incremental learning (TSCIL) remains underexplored due to the scarcity of large-scale time series pre-trained models. Prompted by the recent emergence of large-scale pretrained models (PTMs) for time series data, we present the first exploration of PTM-based Time Series Class-Incremental Learning (TSCIL). Our approach leverages frozen PTM backbones coupled with incrementally tuning the shared adapter, preserving generalization capabilities while mitigating feature drift through knowledge distillation. Furthermore, we introduce a Feature Drift Compensation Network (DCN), designed with a novel two-stage training strategy to precisely model feature space transformations across incremental tasks. This allows for accurate projection of old class prototypes into the new feature space. By employing DCN-corrected prototypes, we effectively enhance the unified classifier retraining, mitigating model feature drift and alleviating catastrophic forgetting. Extensive experiments on five real-world datasets demonstrate state-of-the-art performance, with our method yielding final accuracy gains of 1.4%–6.1% across all datasets compared to existing PTM-based approaches. Our work establishes a new paradigm for TSCIL, providing insights into stability-plasticity optimization for continual learning systems. Copyright © 2025, The Authors. All rights reserved.

关键词： Time series

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：