检索结果-内蒙古大学图书馆

Emerging Smart Computing and Informatics (ESCI), Conference on

作者： Punyala Vijaya Lakshmi Prakash Vasanth B Santhosh Kumar Krishnamoorthy. J Poovannan. E C. Rajesh Department of Computer Science and Engineering Saveetha Engineering College Chennai India Department of Computer Science and Engineering ( Internet of Things) Saveetha Engineering College Chennai India Department of Electronics and Communication Engineering Saveetha Engineering College Chennai India

ISBN: (数字)9798331515683

ISBN: (纸本)9798331515690

By improving the precision of genetic abnormality diagnosis and prediction, machine learning is transforming healthcare and creating opportunities for personalized treatments and early intervention. This research explores the application of machine learning methodologies in the analysis of patient medical records and large genomic datasets for genetic disease prediction. The technology correctly diagnoses hereditary disorders such as Huntington’s disease, muscular dystrophy, and cystic fibrosis through complex approaches such as neural networks, decision trees, and support vector machines. By integrating crucial elements such as genetic mutations, family medical history, and environmental factors, the framework efficiently evaluates possible genetic hazards. This creative approach facilitates the implementation of individualized treatments and preventive healthcare measures, which reduces the load on medical resources while improving patient care. The project highlights how revolutionary data-driven.

关键词： Support vector machines Accuracy Neural networks Machine learning Medical services Genetics Data models Robustness Decision trees Medical diagnostic imaging

来源：评论

学校读者我要写书评

暂无评论

Segmentation of Brain Tumor Magnetic Resonance Imaging using Convolutional Neural Networks 3

Segmentation of Brain Tumor Magnetic Resonance Imaging using...

引用

3rd International Conference on Intelligent Systems, Advanced Computing, and Communication, ISACC 2025

作者： Gupta, Anuj Anita Jaypee University of Information Technology Waknaghat Department of Computer Science & Engineering and Information Technology H.P. Solan India

ISBN: (纸本)9798331523893

In today's world, one of the most aggressive and common diseases that in its greatest rise leads to short life expectancy is known to be brain tumors. In the current work, for diagnosing brain tumors, we have used MRI images which generate enormous data which at a particular time impedes the tumor vs non-tumor's manual classification. Along with some disadvantages of this technique, it has been observed that for a restricted number of images, it provides precise quantitative measurements. Thus, to prevent the human death rate, there is a need for an automatic and trusted classification scheme. Thus, we are proposing CNN classification for the automatic detection of brain tumors in this work. For exploring a small 3×3 kernel, a CNN-based automatic segmentation approach is suggested in this paper. Small kernels are used as they help in deeper architecture design, along with a positive influence against overfitting, as the network consists of a small number of weights. © 2025 IEEE.

关键词： Brain Tumor Segmentation Convolutional Neural Networks Deep Learning Magnetic Resonance Imaging

来源：评论

学校读者我要写书评

暂无评论

Energy-Efficiency Optimization in IoT-Based Machine Learning for Smart Environmental Monitoring 6th

Energy-Efficiency Optimization in IoT-Based Machine Learning...

引用

6th International Conference on Computational Intelligence in Communications and Business Analytics, CICBA 2024

作者： Anand, Nishant Parwekar, Pritee Bali, Vikram Department of Computer Science and Engineering SRM Institute of Science and Technology SRM University Delhi-Meerut Road UP Ghaziabad India Department of Computer Science and Engineering IMS Engineering College UP Ghaziabad India

ISBN: (纸本)9783031813351

The rapid proliferation of the Internet of Things (IoT) technology has paved the way for a more sustainable and efficient approach to environmental monitoring. In this research paper, a comprehensive comparison-based review of energy-efficient strategies in IoT-based machine learning for smart environmental monitoring is presented. This study addresses the need for optimizing energy consumption while maintaining the accuracy and reliability of data collection and analysis in smart environmental monitoring systems. Starting by discussing the evolution of IoT-based environmental monitoring and its significance in the context of creating a smarter and more energy-efficient society. The paper examines a wide range of machine-learning techniques and algorithms commonly employed for environmental data analysis and their respective energy consumption profiles. This findings reveal that, by carefully selecting and implementing energy-efficient machine learning algorithms, it is possible to significantly reduce the power consumption of IoT devices, extending their operational lifespans and reducing the need for frequent battery replacements or recharging. Additionally, the discussion is done on the impact of data compression and transfer techniques, as well as low-power communication protocols, in further enhancing energy efficiency within IoT systems. Also delve into various case studies and practical implementations, highlighting successful approaches and their outcomes in terms of energy efficiency and data quality. The paper concludes by emphasizing the critical importance of continuous research and development in the realm of energy-efficient IoT-based machine learning for smart environmental monitoring, to achieve a more sustainable and eco-friendly smart society. © The Author(s), under exclusive license to Springer Nature Switzerland AG 2025.

关键词： Environmental monitoring

来源：评论

学校读者我要写书评

暂无评论

Encoding laparoscopic image to words using vision transformer for distortion classification and ranking in laparoscopic videos

引用

Multimedia Tools and Applications 2025年第10期84卷 7159-7181页

作者： AlDahoul, Nouar Karim, Hezerul Abdul Momo, Mhd Adel Tan, Myles Joshua Toledo Fermin, Jamie Ledesma Faculty of Engineering Multimedia University Cyberjaya63100 Malaysia Computer Science New York University Abu Dhabi United Arab Emirates College of Engineering and Technology University of St. La Salle Bacolod6100 Philippines Department of Natural Sciences University of St. La Salle Bacolod6100 Philippines

Laparoscopic videos are tools used by surgeons to insert narrow tubes into the abdomen and keep the skin without large incisions. The videos captured by a camera are prone to numerous distortions such as uneven illumination, motion blur, defocus blur, smoke, and noise which have impact on visual quality. Automatic detection and identification of distortions are significant to enhance the quality of laparoscopic videos to avoid errors during surgery. The video quality assessment includes two stages: classification of distortions affecting the video frames to identify their types and ranking of distortions to estimate the intensity levels. The dataset generated in ICIP2020 challenge including laparoscopic videos was utilized for training, validation, and testing the proposed solution. The difficulty of this dataset is caused by having five categories of distortions and four levels of severity. Additionally, the availability of multiple distortion categories in one video is considered the most challenging part of this dataset. The work presented in this paper contributes to solve the multi-label distortion classification and ranking problem. This paper aims to enhance the performance of distortion classification solutions. Vision transformer which is a deep learning model was used to extract informative features by transferring learning and representation from the general domain to the medical domain (laparoscopic videos). Additionally, six parallel multilayer perceptron (MLP) classifiers were added and attached to vision transformer for distortion classification and ranking. The experiment showed that the proposed solution outperforms existing distortion classification methods in terms of average accuracy (89.7%), average single distortion F1 score (94.18%), and average of both single and multiple distortions F1 score (96.86%). Moreover, it can also rank the distortions with an average accuracy of 79.22% and average F1 score of 78.44%. Hence, the high performance of t

关键词： Classification (of information)

来源：评论

学校读者我要写书评

暂无评论

Efficient Smart Public Transit System Using RFID and IoT integration with MERN web applications

Efficient Smart Public Transit System Using RFID and IoT int...

引用

International Conference on computer, Electrical & Communication engineering (ICCECE)

作者： Kartik Madasanal Anand Gudnavar Sreedevi Nagarmunoli Basavaraj Madagouda Keerti Naregal Department of Computer Science and Engineering Jain College of Engineering and Research Visvesaraya Technological University Belagavi India Department of Computer Science and Engineering Graphic Era University Dehradun India Department of Computer Science and Engineering S.G.Balekundri Institute of Technology Visvesaraya Technological University Belagavi India

ISBN: (数字)9798331513894

ISBN: (纸本)9798331513900

This paper develops a Smart Public Transportation System using RFID technology, IoT integration, and a MERN-based web application for increasing the efficiency, accessibility, and user experience of urban transit systems. It addresses problems such as unreliable bus tracking, lack of support for multilingualism, and ineffective communication between operators and passengers. The key hardware components include RFID tags for bus identification, and IoT-enabled bus stations with RFID scanners, speakers, GSM modules for data transmission, and LCD displays. These ensure accurate real-time updates about the bus locations. On the software side, it is supported by a MERN stack, including MongoDB, ***, ***, and ***, providing a responsive web interface that’s complemented by *** for dynamic route visualization and WebSockets for live updates. Multilingual support through i18next makes it inclusive for various user groups. The system is efficient in that it only updates when RFID scanners capture bus tags at certain checkpoints, thereby reducing server load much more than GPS-based systems. Additional features include route tracking, interactive maps, ticket pricing, and anonymous passenger chats. Crowd-sourced feedback mechanisms and an admin panel for bus management ensure the reliability and scalability of the system. This work shows the feasibility of cost-effective, scalable, and user-centric modernization of public transportation systems, paving the way for further enhancements such as integrated payment systems and AI-based route optimization.

关键词： GSM Real-time systems User experience Software Multilingual Internet of Things Servers Reliability Public transportation Radiofrequency identification

来源：评论

学校读者我要写书评

暂无评论

FedHybrid: Unifying Aggregation Strategies to Optimize Federated Learning on Non-IID Dataset

引用

Procedia computer science 2025年 258卷 3126-3134页

作者： Nagireddy Moneesh Madhan S Divya G. Nair Jyothisha J. Nair Department of Computer Science and Engineering Amrita School of Computing Amrita Vishwa Vidyapeetham Amritapuri-690546 India

Federated Learning (FL) allows the execution of Machine Learning at each device while keeping the data at that device. One of the key research difficulties in FL is the development of good aggregation schemes especially when the data is non-IID. This paper introduces a novel aggregation technique, FedHybrid, which combines the strengths of three existing methods:, FedAvg, FedProx, and FedScafold. The proposed method is referred to as FedHybrid since it utilizes model averaging from FedAvg, a proximal term from FedProx to cope with data heterogeneity, and control variates from FedScaffold to minimize update variance. The results of FedHybrid have been tested on the MNIST/CIFAR-10 datasets in a FL environment of 100 clients and over 100 rounds. These experiments show that FedHybrid enhances the overall accuracy and the rate of convergence compared to FedAvg, FedProx, and FedScafold. In particular, the proposed method, obtained 94.12% of accuracy in the MNIST and 93.52% in CIFAR-10 datasets surpassing the other techniques. The improved performance of our proposed model in non-IID data scenarios implies that this work is another valuable addition to the implementable list of federated learning algorithms, promising higher convergence speed and expressly higher accuracy when confronted with commonly realistic non-IID data distributions.

关键词： Federated Learning FedAvg FedProx FedScaffold FedHybrid Non-IID Data Machine Learning Model Averaging Proximal Term Control Variates

来源：评论

学校读者我要写书评

暂无评论

Cutting-edge approaches in intrusion detection systems: a systematic review of deep learning, reinforcement learning, and ensemble techniques

引用

Iran Journal of computer science 2025年 1-31页

作者： Kalpani, Nethma Rodrigo, Nureka Seneviratne, Dilmi Ariyadasa, Subhash Senanayake, Janaka Department of Computer Science and Informatics Uva Wellassa University Badulla Sri Lanka School of Computing Engineering and Technology Robert Gordon University Aberdeen AB10 7GJ United Kingdom

The growing number of networked devices and complex network infrastructures necessitates robust network security measures. Network intrusion detection systems are crucial for identifying and mitigating malicious activities within network environments. Traditional intrusion detection systems (IDS) often struggle to adapt to new and evolving threats. To address these limitations, researchers are increasingly turning to advanced methodologies such as deep learning (DL), reinforcement learning (RL), and ensemble learning (EL), which offer enhanced capabilities in detecting sophisticated and previously unseen attacks. This systematic literature review (SLR) evaluates 33 technical studies from 2020 to 2024 on the application of DL, RL, and EL in IDS. The study reviews the DL and RL methods used in IDS, along with detection methods that employ EL techniques, to combine the strengths of these approaches. It highlights the advantages, disadvantages, and applicability of the proposed techniques, while also suggesting potential improvements. The aim of this SLR is to provide researchers and practitioners with a comprehensive understanding of current IDS methodologies leveraging DL, RL, and EL, enabling the development of more effective and resilient network security solutions. In addition, this review identifies gaps in the existing literature and outlines potential future research directions. © The Author(s), under exclusive licence to Springer Nature Switzerland AG 2025.

关键词： Deep learning Ensemble learning Intrusion detection systems Network security Reinforcement learning

来源：评论

学校读者我要写书评

暂无评论

Trusted Crowd Funding Platform: Using Blockchain for Funders and Fundraisers

Trusted Crowd Funding Platform: Using Blockchain for Funders...

引用

International Conference on Inventive Computation Technologies (ICICT)

作者： Syed Rabiya M Abisha D Ida Christy J Navedha Evanjalin R Amutha K Department of Computer Science and Engineering Vel Tech Rangarajan Dr. Sagunthala R&D Institute of Science and Technology Chennai Department of Computer Science and Engineering National Engineering College Kovilpatti India

ISBN: (数字)9798331512248

ISBN: (纸本)9798331512255

Crowdfunding has gained popularity as an innovative approach to raise capital for startups, social initiatives, and creative ventures. However, traditional centralized crowdfunding platforms often face critical challenges such as lack of transparency, risk of fraud, and high intermediary costs. To address these limitations, this paper proposes a decentralized crowdfunding platform built on the Ethereum blockchain, leveraging smart contract technology to ensure transparency, accountability, and security. The primary objective of this work is to design a trustworthy and automated system where funders and fundraisers can interact directly without relying on third-party intermediaries. The platform enforces pre-defined conditions for fund distribution, allowing profit-sharing and fund release mechanisms to be executed autonomously through smart contracts. This enhances both the reliability and precision of transactions. Funders have real-time visibility into project progress, improving their confidence and decision-making. The proposed solution also reduces operational costs by automating core functionalities like funding, profit allocation, and capital release. All interactions are securely recorded on the blockchain, ensuring auditability and eliminating manipulation. This paper presents the system's architecture, implementation strategy, and evaluation results, demonstrating the effectiveness of the blockchain-based model in overcoming the shortcomings of centralized platforms. The proposed platform offers a more secure, efficient, and transparent alternative for the crowdfunding ecosystem.

关键词： Crowdfunding Costs Smart contracts Real-time systems Blockchains Fraud Security Resource management Reliability Faces

来源：评论

学校读者我要写书评

暂无评论

Weighted layer aggregation with fast and local label expanding method for community detection in multiplex networks

引用

Physica A: Statistical Mechanics and its Applications 2025年 670卷

作者： Shahgholi, Pouya Bouyer, Asgarali Arasteh, Bahman Liu, Xiaoyang Department of software engineering Azarbaijan Shahid Madani University Tabriz Iran Department of software engineering Faculty of Engineering and Natural Science Istinye University İstanbul Turkey Department of Computer Science Khazar University Baku Azerbaijan School of Computer Science and Engineering Chongqing University of Technology Chongqing400054 China

Community detection in multiplex networks has emerged as a crucial research area due to its ability to capture complex interactions across multiple layers of interconnected data. Despite significant advancements, existing methods often face critical challenges, including computational time, resolution limit, free parameter tuning, training models, etc. To overcome these limitations, this paper presents LCDMN (Layer-Coupled Diffusion for Multiplex Networks) algorithm designed for accurate and efficient community detection in multiplex networks. LCDMN employs dynamic scaling and layer coupling to adaptively identify community structures across diverse network configurations, offering improved resilience to network's density and structural ambiguity. LCDMN addresses the challenges of layer diversity by: (1) dynamically weighting layers based on critical parameters such as layer correlation, layer nodes activity variance, and attractiveness, (2) developing a robust node scoring method, (3) the aggregating layers of multiplex network into a single-layer, weighted graph, (4) employing a label diffusion approach with mechanisms for handling overlapping nodes, and (5) refining community structures through a dynamic merging process that adaptively adjusts layer contributions and community boundaries during execution, ensuring context-sensitive resolution of structural ambiguity. Nodes and edges are scored using network topology and structural metrics to efficiently incorporate in label diffusion process for detecting initial communities. The approach balances computational efficiency with precision, enabling the detection of cohesive and well-defined communities in complex networks. Experimental evaluations on real-world and synthetic multiplex networks demonstrate that LCDMN consistently outperforms state-of-the-art methods, such as Infomap, MDLPA, MPBTV, LART, DGFM3 and GenLouvain, in terms of Normalized Mutual Information (NMI), Adjusted Rand Index (ARI), and modularity.

关键词： Community detection Label diffusion Layer aggregation Multiplex networks Similarity

来源：评论

学校读者我要写书评

暂无评论

Indian legal text summarization using large language model

引用

Procedia computer science 2025年 259卷 1398-1406页

作者： Vishnu Kesav Omanakuttan Rohith A Uthara R Geetha M Department of Computer Science and Engineering Amrita School of Computing Amrita Vishwa Vidyapeetham Kollam India 690546

This research introduces a novel framework for summarizing legal judgments by integrating two state-of-the-art pretrained models—Google T5-small and Facebook BART—with rhetorical labels to enhance both accuracy and relevance. Each model is fine-tuned to pro- cess distinct rhetorical components, such as facts, arguments, and conclusions, while LoRA adapters ensure efficient fine-tuning with minimal computational over- head. The incorporation of rhetorical labels allows the models to differentiate between various legal roles, ensuring the generated summaries preserve the logical structure and critical reasoning embedded in the text. Our experiments demonstrate that this multi- model, label-guided approach not only outperforms traditional summa- rization methods but also delivers coherent, context aware summaries tailored to the needs of legal practitioners, improving both usability and decision-making efficiency.

关键词： Legal judgment summarization pretrained models Google T5-small Facebook BART rhetorical labels fine-tuning LoRA adapters facts arguments conclusions computational efficiency multi-model approach label-guided summarization

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：