检索结果-内蒙古大学图书馆

Detection of COVID-19 Using Machine Learning 1

International Conference on computer Vision and Machine Intelligence, CVMI 2022

作者： Kumar, Saurav Tripathi, Rohit Department of Computer Science and Engineering Indian Institute of Information Technology Guwahati India

ISBN: (数字)9789811978678

ISBN: (纸本)9789811978661

In this paper, we aimed at experimenting and developing suitable machine learning algorithms along with some deep learning architectures, to achieve the task of COVID-19 classification. We experimented with various supervised learning algorithms for the classification of COVID-19 images and normal images, from X-ray images and radiography images. We evaluated the performance of our models on performance metrics like accuracy, precision, recall, F1 score, and Cohen Kappa score. The proposed model achieved the accuracy of 98.5%, F1 score of 97%, precision of 95%, and Cohen Kappa score of 0.88. In this experiment, we jotted down the values of performance of various performance metrics, that were obtained after tuning various parameters of the model and on different image resolutions of input image. We performed linear regression on the results to study the behavior of various performance metrics when we tuned our model with different parameters with different input image resolutions. © 2023, The Author(s), under exclusive license to Springer Nature Singapore Pte Ltd.

关键词： COVID-19

来源：评论

学校读者我要写书评

暂无评论

Leaf Spot Disease Severity Measurement in Terminalia Arjuna Using Optimized Superpixels 7th

Leaf Spot Disease Severity Measurement in Terminalia Arjuna ...

引用

7th International Conference on computer Vision and Image Processing, CVIP 2022

作者： Samanta, Sourav Pratihar, Sanjoy Chatterji, Sanjay Computer Science and Engineering Indian Institute of Information Technology Kalyani741235 India

ISBN: (纸本)9783031314162

Early diagnosis of plant leaf disease, i.e., detection in the initial development stage, is a promising area of research focusing on smart agriculture involving computer vision. Automatic detection can significantly minimize human labor employment due to regular supervision. Terminalia Arjuna is a multi-purpose tree primarily found on the indian subcontinent. The various chemical compounds of the Arjuna leaf are frequently used in medicine. Also, Terminalia Arjuna leaf is utilized in sericulture as a food source for moths. Leaf spot disease of Arjuna is common, and it is necessary to initiate treatment as soon as the disease appears on the leaves to prevent further spread on other leaves and other trees. This study used a multi-objective optimized simple linear iterative clustering (SLIC) algorithm to precisely locate leaf spots in affected areas. The entire leaf surface is segmented into two types of superpixels, healthy and unhealthy. The color moment features have been extracted for classification. The classification accuracy of two types of superpixels using four well-known classifiers has been reported, and SVM achieved the highest classification accuracy at 99.60%. Based on the categorized superpixels, the severity score of the leaf spot disease has been computed for various sample leaves. The experimental findings demonstrate the applicability and robustness of the proposed method. © The Author(s), under exclusive license to Springer Nature Switzerland AG 2023.

关键词： Early detection of leaf spot Multi-objective cuckoo search Simple Linear Iterative Clustering (SLIC) Smart sericulture Terminalia Arjuna leaf

来源：评论

学校读者我要写书评

暂无评论

SlidEar: Exploring eSense in-Ear Wearable for Voice-Assisted Smart Slide Supervision 17

SlidEar: Exploring eSense in-Ear Wearable for Voice-Assisted...

引用

17th International Conference on COMmunication Systems and NETworkS, COMSNETS 2025

作者： Mondal, Susmita Zingre, Sameeran Ravishankar Chakraborty, Suchetana Indian Institute of Technology Department of Computer Science and Engineering Jodhpur India

ISBN: (纸本)9798331531195

In corporate settings, conferences, or classrooms, an orator relies on manual slide transitions, which can disrupt their presentation flow. Presentation devices can be inaccessible due to the physical limitations of individuals, reducing engagement with the audience. Traditionally used remote-controlled wireless presenters are hand-held devices that usually require a compatible connector slot for the presentation device and partially engage one hand of the orator. To ease the hand-free presentation of slides, we introduce SlidEar, a voice-assisted smart slide supervision successfully tested on eSense in-ear wearables. SlidEar leverages voice recognition fusion with wearable sensors to facilitate seamless slide operations. It allows real-time feedback on natural language commands during the presentation of slides. Compared with conventional methods, SlidEar enhances engagement and accessibility through hand-free in-ear placement and helps the orator focus on delivering their content without any disruption or distraction. The evaluation shows promising results that improve the presentation experience profusion. SlidEar significantly reduces errors and can be widely adopted in diverse presentation scenarios. © 2025 IEEE.

关键词： Speech recognition

来源：评论

学校读者我要写书评

暂无评论

OSINT at GenAI Detection Task 1: Multilingual MGT Detection: Leveraging Cross-Lingual Adaptation for Robust LLMs Text Identification 1

OSINT at GenAI Detection Task 1: Multilingual MGT Detection:...

引用

1st Workshop on GenAI Content Detection, GenAIDetect 2025

作者： Agrahari, Shifali Singh, Sanasam Ranbir Department of Computer Science and Engineering Indian Institute of Technology Guwahati India

ISBN: (纸本)9798891762053

Detecting AI-generated text has become increasingly prominent. This paper presents our solution for the DAIGenC Task 1 Subtask 2, where we address the challenge of distinguishing human-authored text from machine-generated content, especially in multilingual contexts. We introduce Multi-Task Detection (MLDet), a model that leverages Cross-Lingual Adaptation and Model Generalization strategies for Multilingual Machine-Generated Text (MGT) detection. By combining language-specific embeddings with fusion techniques, MLDet creates a unified, language-agnostic feature representation, enhancing its ability to generalize across diverse languages and models. Our approach demonstrates strong performance, achieving macro and micro F1 scores of 0.7067 and 0.7187, respectively, and ranking 15th in the competition1. We also evaluate our model across datasets generated by different distinct models in many languages, showcasing its robustness in multilingual and cross-model scenarios. © 2025 International Conference on Computational Linguistics.

关键词： Computational linguistics

来源：评论

学校读者我要写书评

暂无评论

Accelerating Native Inference Model Performance in Edge Devices using TensorRT 6

Accelerating Native Inference Model Performance in Edge Devi...

引用

6th IEEE International Conference on Recent Advances in Intelligent Computational Systems, RAICS 2024

作者： Gopiktrishna, P.B. Marion Lincy G, Rubell Rishekeeshan, Abhishek Deekshitha Indian Institute of Information Technology Computer Science and Engineering Kottayam India AiRender Technology Pvt Ltd Bangalore India

ISBN: (纸本)9798350381689

In this era, deep learning is becoming increasingly popular for solving real-world problems. Due to the extremely high processing power required to execute the most complex deep learning models, Graphics Processing Units(GPUs) and Neural Processing Units(NPUs) are essential for optimal execution. However, it is challenging to incorporate these high-performance processors in every edge device like Internet of Things gadgets, embedded systems, and cell phones to execute complex inference models. Hence, deep learning models must be optimized to perform well on edge devices with limited memory, power, and computing capabilities. Optimizing the inference models to accelerate performance on edge devices is one of the significant issues facing, especially in complex computation domains like computer vision. TensorRT, introduced by NVIDIA, is one of the best approaches used to optimize inference models, by accelerating their performance on edge devices without compromising the model accuracy. It presents a variety of optimization approaches, including layer fusion, quantization, dynamic tensor memory management, kernel auto tuning, and precision calibration. Every optimization technique is contingent upon the problem that have selected and the use of all optimization approaches has the potential to significantly speed up the performance of an inference model. This paper explores how TensorRT accelerates the performance of inference models like ResNet-50, VGG16, and Inception V3 in an image classification problem, and the results are analyzed with its native inference models. The utilization of TensorRT in edge devices is expanding significantly, reflecting a growing scope for enhanced performance and efficiency in processing complex neural networks at the edge. © 2024 IEEE.

关键词： Graphics processing unit

来源：评论

学校读者我要写书评

暂无评论

Data-driven output regulation control for constrained linear systems

引用

science China(information sciences) 2025年第3期68卷 338-353页

作者： Chaoyu XIA Yi DONG Chaoli WANG Shengyuan XU Shanghai Research Institute for Intelligent Autonomous Systems Tongji University College of Electronic and Information Engineering Shanghai Research Institute for Intelligent Autonomous SystemsTongji University Department of Control Science and Engineering School of Optical-Electrical and Computer EngineeringUniversity of Shanghai for Science and Technology Department of Automation Nanjing University of Science and Technology

This study introduces a data-driven approach for state and output feedback control addressing the constrained output regulation problem in unknown linear discrete-time systems. Our method ensures effective tracking performance while satisfying the state and input constraints, even when system matrices are not available. We first establish a sufficient condition necessary for the existence of a solution pair to the regulator equation and propose a data-based approach to obtain the feedforward and feedback control gains for state feedback control using linear programming. Furthermore, we design a refined Luenberger observer to accurately estimate the system state, while keeping the estimation error within a predefined set. By combining output regulation theory, we develop an output feedback control strategy. The stability of the closed-loop system is rigorously proved to be asymptotically stable by further leveraging the concept of λ-contractive sets.

关键词： output regulation constrained system data-driven.

来源：评论

学校读者我要写书评

暂无评论

Learning Non-cooperative Driving Practices and its Impact on Road Traffic Dynamics 17

Learning Non-cooperative Driving Practices and its Impact on...

引用

17th International Conference on COMmunication Systems and NETworkS, COMSNETS 2025

作者： Osho Chakraborty, Suchetana Indian Institute of Technology Department of Computer Science and Engineering Jodhpur India

ISBN: (纸本)9798331531195

Cooperative driving that prioritizes global safety and efficiency over self-interest is crucial in ensuring smooth traffic flow and enhanced road safety. However, experiencing noncooperative driving is common under heterogeneous and chaotic on-road traffic conditions. While non-cooperativeness is often context-dependent and assessed during direct vehicle interactions, it can be inferred through longitudinal analysis of driving patterns from vehicular kinematics data. This behavior not only adversely affects the vehicle itself but also has a cascading impact on nearby vehicles, ultimately disrupting overall traffic flow. In this study, we quantify a vehicle's non-cooperativeness using kinematic data and further model its cascading effects on nearby cooperative vehicles. Experimental analysis was conducted using the SUMO traffic simulator and the TraCI interface to assess the impact of non-cooperativeness on overall traffic dynamics. This research is among the first to examine the cascading effects of a vehicle's non-cooperativeness on its cooperative nearby vehicles. © 2025 IEEE.

关键词： Motor transportation

来源：评论

学校读者我要写书评

暂无评论

TruChain: A Blockchain-based Access Control to Improve the Security of Smart Water Grid Systems 17

TruChain: A Blockchain-based Access Control to Improve the S...

引用

17th International Conference on COMmunication Systems and NETworkS, COMSNETS 2025

作者： Mondal, Susmita Chakraborty, Suchetana Indian Institute of Technology Department of Computer Science and Engineering Jodhpur India

ISBN: (纸本)9798331531195

Water is a significant resource in day-to-day life, and it usually requires technological association for comprehensive management. Smart Water Grids (SWG) typically use cyberphysical systems (CPS) to monitor several parameters associated with water quality and distribution, leveraging the possibility of performing remote operations. For robust architecture, remote operations rely on application-specific sensors, communication networks, edge control, and cloud computation. Needless to say, the centralized control of edge and cloud infrastructure is prone to attacks and failures, such as unauthorized access or single-point-of-failure. We propose TruChain, a Blockchain-based access control mechanism to eliminate the probability of access takeover and aggregate quorum-based validation on sensory data for transparent provenance reliability. We also provide access revocation to strengthen security by restricting access to sensitive data stored in a privately distributed cloud. We believe our solution has empirical decentralization and a verifiable authorization method by maintaining a tamper-evident, auditable trail of interactive events, building trust without third-party intervention. © 2025 IEEE.

关键词： Cloud security

来源：评论

学校读者我要写书评

暂无评论

Computational Politeness in Natural Language Processing: A Survey

引用

ACM Computing Surveys 2024年第9期56卷 1页

作者： Priya, Priyanshu Firdaus, Mauajama Ekbal, Asif Computer Science and Engineering Indian Institute of Technology Patna Bihar Patna India University of Alberta EdmontonAB Canada

Computational approach to politeness is the task of automatically predicting and/or generating politeness in text. This is a pivotal task for conversational analysis, given the ubiquity and challenges of politeness in interactions. The computational approach to politeness has witnessed great interest from the conversational analysis community. This article is a compilation of past works in computational politeness in natural language processing. We view four milestones in the research so far, viz. supervised and weakly supervised feature extraction to identify and induce politeness in a given text, incorporation of context beyond the target text, study of politeness across different social factors, and study the relationship between politeness and various socio-linguistic cues. In this article, we describe the datasets, approaches, trends, and issues in computational politeness research. We also discuss representative performance values and provide pointers to future works, as given in the prior works. In terms of resources to understand the state of the art, this survey presents several valuable illustrations - most prominently, a table summarizing the past papers along different dimensions, such as the types of features, annotation techniques, and datasets used. © 2024 Copyright held by the owner/author(s). Publication rights licensed to ACM.

关键词： Computational methods

来源：评论

学校读者我要写书评

暂无评论

QuMIN: quantum multi-modal data fusion for humor detection

引用

Multimedia Tools and Applications 2025年第18期84卷 18855-18872页

作者： Phukan, Arpan Haq Khan, Anas Anwarul Ekbal, Asif Department of Computer Science and Engineering Indian Institute of Technology Patna Bihta Bihar Patna801106 India Kalinga Institute of Industrial Technology Odisha Bhubaneswar751024 India School of AI and Data Science Indian Institute of Technology Jodhpur Rajasthan Jodhpur342030 India

Humour detection has attracted considerable attention due to its significance in interpreting dialogues across text, visual, and acoustic modalities. However, effective methods to map correlations among different modalities remain an active area of research. In this study, we go beyond traditional machine learning techniques by introducing a Variational Quantum Circuit (VQC) that capitalizes on the inherent quantum properties of superposition, entanglement, and interference. Our proposed model, Quantum Multi-Modal Data Fusion (QuMIN), is designed to better capture and reproduce the interaction across modalities, as well as the internal correlations within each modality. Our introduction of the novel VQC, which augments the DialogueRNN baseline with only an additional 4,809 parameters, signifies a substantial advancement in multi-modal humor detection with improvements of 12.34% in precision, 8.84% in recall and 10.57% in F1 score compared to the state-of-the-art methods. © The Author(s), under exclusive licence to Springer science+Business Media, LLC, part of Springer Nature 2024.

关键词： Machine 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：