检索结果-内蒙古大学图书馆

Comprehensive Earth System Analysis for Advancing Community Resilience: Integrating Remote Sensing with Hybrid GAN-RNN Techniques

引用

Remote Sensing in Earth Systems Sciences 2024年第4期7卷 788-803页

作者： Sundar, Ganapathy Patchaiammal, Pitchaimuthu Lakkshmanan, Ajanthaa Ramya, Saminathan Pallavi, Lanke Thiagarajan, Ranganathan Department of Computer Applications Sindhi College Chennai India Department of Computing Technologies SRM Institute of Science and Technology Kattankaluthur Chengalpattu District India Department of Computer Science and Engineering Kongu Engineering College Perundurai Tamil Nadu Erode India Department of CSE B V Raju Institute Of Technology Narsapur India Department of Information Technology Prathyusha Engineering College Thiruvallur India

Effective disaster prediction is essential for climate resilience, especially in areas prone to extreme weather events. Conventional models often face challenges in predicting complex or uncommon events, largely due to data scarcity and the difficulty of modeling temporal dependencies in sequential data. This paper introduces an innovative hybrid GAN-RNN model that leverages remote sensing and IoT sensor data to improve disaster prediction accuracy. By combining generative adversarial networks (GANs) for generating synthetic data and recurrent neural networks (RNNs) for recognizing temporal patterns, this model forms a potent predictive tool. Achieving a remarkable testing accuracy of 98.16%, the hybrid GAN-RNN model significantly surpasses traditional predictive methods. Comprehensive evaluations demonstrate its enhanced capability to predict complex and rare events effectively, establishing it as a crucial resource for disaster risk management and environmental monitoring. This research advances AI-driven methodologies in climate resilience, offering a solid framework for future innovations in the field. © The Author(s), under exclusive licence to Springer Nature Switzerland AG 2024.

关键词： AI-driven solutions Climate resilience Data augmentation Disaster prediction Environmental monitoring Hybrid GAN-RNN IoT sensors Remote sensing Synthetic data Temporal pattern recognition

来源：评论

学校读者我要写书评

暂无评论

Leveraging AI to Unlock the Potential of Data Science 15

Leveraging AI to Unlock the Potential of Data Science

引用

15th International Conference on Computing Communication and Networking Technologies, ICCCNT 2024

作者： Yadav, Rakesh Kumar Sudhamsu, Gadug Goyal, Manish Kumar Yuvaraj, S. Deshpande, Leena A. Maharishi University of Information Technology Maharishi School of Engineering and Technology Uttar Pradesh India Faculty of Engineering and Technology Department of Computer Science and Engineering Karnataka Bangalore India Vivekananda Global University Department of Computer Science & Application Jaipur India Prince Shri Venkateshwara Padmavathy Engineering College Department of Electronics and Communication Engineering Chennai127 India Vishwakarma Institute of Technology Department of Computer Engineering Pune India

ISBN: (纸本)9798350370249

The application of synthetic Intelligence (AI) to records science has spread out a massive array of new possibilities. Via the combination of advanced evaluation strategies, effective understanding-pushed system gaining knowledge of algorithms, and modern computing architectures, AI is allowing researchers to quick and as it should be discover insights from huge and complicated datasets. Furthermore, AI tools are offering corporations with the ability to make higher and more knowledgeable choices from the evaluation of hundreds of thousands of statistics points. The skills of information technology powered by means of AI are revolutionizing industries which includes healthcare, finance, and manufacturing-producing savings, uncovering efficiencies, and identifying new opportunities. As AI-driven records technology maintains to strengthen, organizations that capitalize on its ability stand to gain precious insights and unlock new degrees of overall performance. © 2024 IEEE.

关键词： Smart manufacturing

来源：评论

学校读者我要写书评

暂无评论

Optimizing land mine detection across diverse mining environments: A hyperspectral data approach with regression models

引用

International Journal of Intelligent Networks 2024年第1期5卷 351-363页

作者： Anand, R. Andrew, J. Makki, Ihab Department of Electrical and Electronics Engineering Amrita School of Engineering Amrita Vishwa Vidyapeetham Coimbatore India Department of Computer Science and Engineering Manipal Institute of Technology Manipal Academy of Higher Education Manipal Karnataka 576104 India Department of Nouvelle Aquitaine France

The detection of landmines, namely anti-tank mines, explosive devices, and unexploded ordnance, is a formidable obstacle for the global community. The visible consequences of unobserved explosives in communities affected by war are characterized by significant devastation and human suffering. In order to effectively tackle this matter, it is imperative to use proactive strategies that focus on the identification and mitigation of these perilous substances prior to their potential infliction of harm. Nevertheless, the majority of current solutions exhibit significant deficiencies, such as exorbitant expenses, inefficiencies, and apprehensions over accuracy. These drawbacks are further compounded by the inherent trade-offs that exist between these elements, where improvements in one area often come at the expense of another. Contrarily, recent breakthroughs in the areas of deep learning, unmanned aerial vehicles, and sensor technologies are being recognized as potentially transformative elements in the domain of landmine identification and removal. This paper presents a thorough examination of recent scholarly investigations that integrate computerized technology in the field of landmine detection. To the extent of our current understanding, there has been no prior investigation that has thoroughly examined this particular domain. The main aim of this study is to investigate the incorporation of machine learning based regression methods in the field of landmine detection. The study specifically emphasizes the identification and resolution of existing issues that hinder the development of efficient automated solutions, hence enhancing performance optimization. The Sum of Sine Curve Fit Regression Model is proposed and proved a powerful and adaptable tool for extracting relevant information from this hyperspectral images. © 2024 The Authors

关键词： Unmanned aerial vehicles (UAV)

来源：评论

学校读者我要写书评

暂无评论

Cybersecurity Threat Detection Using Deep Learning and Anomaly Detection Techniques

Cybersecurity Threat Detection Using Deep Learning and Anoma...

引用

2024 International Conference on Knowledge engineering and Communication Systems, ICKECS 2024

作者： Rajendran, T. Mohamed Imtiaz, N. Jagadeesh, K. Sampathkumar, Balaji Rajalakshmi Institute of Technology Department of Computer Science and Engineering Chennai India Nagarjuna College of Engineering and Technology Dept. of Information Science and Engineering Bangalore India Sri Venkateswaraa College of Technology Dept. of Computer Science and Engineering Chennai India Akshaya College of Engineering and Technology Dept. of Computer Science and Engineering Coimbatore India

ISBN: (纸本)9798350359688

Through the use of combined deep learning and anomaly detection approaches, this research investigates the area of cybersecurity threat detection. The study proves the framework's extraordinary success in recognizing as well as mitigating diverse cyber dangers, including insider threats and zero-day attacks, through practical testing and case studies. The framework establishes itself as a proactive cybersecurity solution adaptive to the changing threat landscape due to its high accuracy of 95%, low amount of false positives 5%, and quick reaction times. The recommendations cover user training, scalability, regulatory compliance, including cross-disciplinary collaboration. Behavioral analysis, AI explainability, IoT security, and countering quantum computing concerns should be the main areas of future research. These revelations help the cybersecurity industry become more resilient and well-prepared. In this research investigation, experiments on threat detection were conducted using deep learning algorithms, with a specific focus on Convolutional Neural Networks (CNNs) and Recurrent Neural Networks (RNNs). The findings of the study reveal that a detection accuracy of 93% was achieved by CNN, while a higher accuracy of 96% in threat detection was exhibited by RNN. © 2024 IEEE.

关键词： Zero-day attack

来源：评论

学校读者我要写书评

暂无评论

IIITDWD@LT-EDI-2023 Unveiling Depression: Using pre-trained language models for harnessing domain-specific features and context information 3

IIITDWD@LT-EDI-2023 Unveiling Depression: Using pre-trained ...

引用

3rd Workshop on Language technology for Equality, Diversity and Inclusion, LTEDI 2023

作者： Biradar, Shankar Saumya, Sunil Kavatagi, Sanjana Department of Computer Science and Engineering Indian Institute of Information Technology Karnatka Dharwad India Department of Computer Science and Engineering VTU Belagavi India

ISBN: (纸本)9789544520847

Depression is a global health crisis affecting millions. Workplace stress and unhealthy habits have risen, leading to more people with depressive symptoms. Early detection and prediction of depression are essential for timely intervention and support. Unfortunately, social stigma prevents many from seeking help, making early detection difficult. Therefore, alternative strategies for depression prediction, such as analysing social media posts, are being explored. LT-EDI@RANLP held a shared task to promote research in this field. Our team participated in the shared task and secured 21st rank with a macro F1 score of 0.36. This article summarises the model used in the shared task. © 2023 LTEDI 2023 - 3rd Workshop on Language technology for Equality, Diversity and Inclusion, associated with the 14th International Conference on Recent Advances in Natural Language Processing, RANLP 2023 - Proceedings. All rights reserved.

关键词： Natural language processing systems

来源：评论

学校读者我要写书评

暂无评论

Hybrid generative regressionbased deep intelligence to predict the risk of chronic disease

引用

International Journal of Intelligent Computing and Cybernetics 2022年第1期15卷 144-164页

作者： Sandeep Kumar Hegde Monica R.Mundada Department of Computer Science and Engineering M.S.Ramaiah Institute of TechnologyBangaloreIndia Department of Computer Science and Engineering NMAM Institute of TechnologyNitteIndia Visvesvaraya Technological University BelagaviIndia

Purpose-Chronic diseases are considered as one of the serious concerns and threats to public health across the *** such as chronic diabetes mellitus(CDM),cardio vasculardisease(CVD)and chronic kidney disease(CKD)are major chronic diseases responsible for millions of *** of these diseases is considered as a risk factor for the other two ***,noteworthy attention is being paid to reduce the risk of these *** amount of medical data is generated in digital form from smart healthcare appliances in the current *** numerous machine learning(ML)algorithms are proposed for the early prediction of chronic diseases,these algorithmic models are neither generalized nor adaptive when the model is imposed on new disease ***,these algorithms have to process a huge amount of disease data iteratively until the model *** limitation may make it difficult forMLmodels to fit and produce imprecise results.A single algorithm may not yield accurate ***,an ensemble of classifiers built from multiple models,that works based on a voting principle has been successfully applied to solve many classification *** purpose of this paper is to make early prediction of chronic diseases using hybrid generative regression based deep intelligence network(HGRDIN)***/methodology/approach-In the proposed paper generative regression(GR)model is used in combination with deep neural network(DNN)for the early prediction of chronic *** GR model will obtain prior knowledge about the labelled data by analyzing the correlation between features and class ***,the weight assignment process of DNN is influenced by the relationship between attributes rather than random *** knowledge obtained through these processes is passed as input to the DNN network for further *** the inference about the input data instances is drawn at the DNN through the GR model,the model is named as

关键词： Deep neural network Generative regression Chronic disease Regularization Maximum likelihood estimation

来源：评论

学校读者我要写书评

暂无评论

Firefighter Skill Advancement through IoT-Enabled Virtual Reality and CNN-Based Training 11

Firefighter Skill Advancement through IoT-Enabled Virtual Re...

引用

11th International Conference on Reliability, Infocom Technologies and Optimization, ICRITO 2024

作者： Asha, P. Muthalakshmi, M. Jeslin, J Gnana Nithisha, J. Vijayakumar, S. Srinivasan, C. Sathyabama Institute of Science and Technology Department of Computer Science and Engineering Tamil Nadu Chennai India Veltech Rangarajan Dr.Sagunthala R&d Institute of Science and Technology Department of Biomedical Engineering Tamil Nadu Chennai India Rmk College of Engineering and Technology Department of Computer Science and Engineering Tamil Nadu Chennai India Srm Institute of Science and Technology Department of Computer Science and Engineering Tamil Nadu Chennai India Department of Biomedical Engineering Tamil Nadu Namakkal India Saveetha School of Engineering Saveetha Institute of Medical and Technical Sciences Saveetha University Department of Computer Science and Engineering Tamil Nadu Chennai India

ISBN: (纸本)9798350350357

To maintain the safety and efficacy of firefighters in various circumstances, modern firefighting necessitates constantly improving skills and training techniques. Utilizing the Internet of Things (IoT), virtual reality (VR), and convolutional neural networks (CNN), this paper details a novel method for training firefighters. The proposed system collects real-time data on ambient variables, equipment state, and firefighter biometrics via integrating IoT sensors into firefighting equipment and training settings. Using this information, it can develop lifelike VR training simulations of difficult and potentially dangerous scenarios. To make the training settings more realistic and malleable, CNN-based algorithms are used to assess the data. The capacity to simulate a wide variety of firefighting situations, customize training difficulty depending on individual and team performance, and provide instant feedback and performance metrics to trainees are all major benefits of this method. The method also allows teachers to check in and evaluate their learners remotely, improving instruction quality. An IoT-enabled VR and CNN-based training technique has shown promising preliminary results in pilot trials, suggesting it might greatly enhance firefighter competence, situational awareness, and decision-making ability. Because of this, it has the potential to completely alter the way firefighters are informed and prepared for the ever-changing dangers users may encounter on the job. © 2024 IEEE.

关键词： Internet of things

来源：评论

学校读者我要写书评

暂无评论

Multiple Device-Based Geo-Position Spoofing Detection in Instant Messaging Platform with Residual Noise Extraction Using DRN 6th

Multiple Device-Based Geo-Position Spoofing Detection in Ins...

引用

6th International Conference on Intelligent Communication, Control and Devices, ICICCD 2024

作者： Koparde, Shweta Mane, Vanita Department of Computer Engineering Ramrao Adik Institute of Technology D Y Patil Deemed to Be University Nerul India

ISBN: (纸本)9789819783281

In social media, location registration is emerging as a trend. Concurrently, the geo-position locations are modified by the criminals based on their needs. Thus, it is necessary to identify the authenticity of geo-position. The existing method failed in detecting the prior information and is effective in small datasets. Hence, a multi-aided deep residual network spoof detect (multi-aided DRN_Spoof detect) is developed for the detection of spoofed images in this research. The multi-aided DRN is utilized for residual noise extraction from the accumulated input image. Then, a fuzzy filter is utilized for camera footprint extraction from the residual noise extracted image, and the process is followed for spoof image. Moreover, the RV coefficient is applied for the fusion of the resultant extracted camera footprints. Later, Neyman similarity is applied to the fusion of geotagged images. At last, the spoofed image is identified from the resultant fused camera footprints and fused geotagged images using Motyka similarity. Further, the analysis shows that the multi-aided DRN_Spoof detect attained a true positive rate (TPR), accuracy, and true negative rate (TNR) of 95.79, 93.79, and 92.68%. © The Author(s), under exclusive license to Springer Nature Singapore Pte Ltd. 2025.

关键词： Fuzzy filters

来源：评论

学校读者我要写书评

暂无评论

Machine Versus Human Intelligence: An Illustration from the CSES Problem Set

Machine Versus Human Intelligence: An Illustration from the ...

引用

2024 International Conference on Artificial Intelligence and Emerging technology, Global AI Summit 2024

作者： Dutta, Anurag Ray, Samrat Kumarasankaralingam, Lakshmanan Kumar, Pijush Kanti Ghosh, Soubhik Kumar, Padmanavan Government College of Engineering & Textile Technology Department of Computer Science Serampore India International Institute of Management Sciences Maharashtra Pune India Saint Joseph University Department of Mathematics Nagaland India Government College of Engineering & Textile Technology Department of Information Technology Serampore India Bengal College of Engineering And Technology Department of Computer Science Durgapur India Kalinga Institute of Industrial Technology Department of Computer Science Odisha Bhubaneswar India

ISBN: (纸本)9798350379716

Following the release of CHATGPT in November 2022, both textual and visual LLMs evolved a long way. Several comments have been made by experts on the expertise and intelligence possessed by these Large Language Models;Amongst all, the contest between Human and Machine Intelligence is that discussion that catches great heights. This research focuses on the analytical and reasoning abilities of LLMs and further compares them to those of Human Intelligence. As a benchmark for the same, the CSES Problem Set, a well-structured set of algorithmic problems, touches a wide spectrum of computational syllabi, like Data Structures, Mathematics, Decision Theory, etc. Our analysis of the contrast in abilities leads to the conclusion that LLMs need a long way to go, and thus LLM researchers rather than modeling more and more visually appealing LLMs (like Sora, Dall-e, etc.) should focus more on the abilities of the existing textual models like CHATGPT. Our research shows that the Machine Intelligence is ≈ 46 % lower than that of the Human Intelligence. © 2024 IEEE.

关键词： Problem oriented languages

来源：评论

学校读者我要写书评

暂无评论

Generative Adversarial Networks (GANs) in the Wild: Applications, Challenges, and Future Prospects 2

Generative Adversarial Networks (GANs) in the Wild: Applicat...

引用

2nd International Conference on Disruptive Technologies, ICDT 2024

作者： Kumar, Rakesh Chaudhary, Pooja Gupta, Shilpi Singhal, Manoj Gla University Department of Computer Engineering & Applications UP Mathura India UP Ghaziabad India Meerut Institute of Technology Department of Computer Science & Engineering UP Meerut India Gl Bajaj Institute of Technology & Management Department of Information Technology UP Greater Noida India

ISBN: (纸本)9798350371055

Generative adverse Networks (GANs) have become increasingly famous for a variety of tasks in picture and audio processing, especially unsupervised getting to know. GANs encompass components, a 'generator' and a 'discriminator,' that are skilled adversarial. The generator is tasked with producing a pattern of statistics, and the discriminator is tasked with figuring out whether the information is 'actual' or 'fake.' The generator 'learns' through trial and mistakes as it attempts to create samples that appear practical to the discriminator. Through the antagonistic schooling manner, the generator can create more accurate 'fake' records while the discriminator can emerge as better at distinguishing between 'actual' and 'faux' records. GANs have been applied to various duties, which include picture era, textual content to photosynthesis, photo-to-image translation, and automatic music composition. Recent tries have been made to apply GANs to extra complicated obligations in the wild, including self-reliant use and drug discovery. GANs have inspired an ever-developing quantity of industries because of their potential to generate sensible facts with minimal human input. © 2024 IEEE.

关键词： antagonistic complicated discriminator photosynthesis potential

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：