检索结果-内蒙古大学图书馆

2024 International Conference on computer, Electronics, Electrical engineering and their Applications, IC2E3 2024

作者： Kancharla, Jaganmohan Reddy Madhu Kumar, S.D. National Institute of Technology Department of Computer Science & Engineering Calicut India

ISBN: (纸本)9798350388534

Data localization in distributed computing has gotten a lot of attention because it has the potential to improve data sovereignty, make sure regulations are followed, and make data easier to access. However, this practice also brings forth a multitude of challenges and threats to cybersecurity, data privacy, and overall information security. In this comprehensive survey paper, we delve into the complex landscape of challenges and threats associated with data localization in distributed computing environments. We explore the intricate interplay between data localization, distributed infrastructure, cybersecurity, and privacy concerns. This paper provides an in-depth analysis of various challenges such as regulatory compliance, data integrity, authentication, and access control, along with potential threats like data breaches, insider attacks, and unauthorized access. Furthermore, we investigate the state-of-the-art security mechanisms, encryption techniques, privacy-preserving methods, and access control strategies that researchers and practitioners have developed to address these challenges. Through this survey, we aim to provide a comprehensive overview of the critical issues surrounding data localization within cloud computing and shed light on the emerging trends and potential solutions that can pave the way for secure and privacy-aware data localization practices. © 2024 IEEE.

关键词： Data integrity

来源：评论

学校读者我要写书评

暂无评论

Machine Learning-Driven Approaches for Dermatological Disease Diagnosis 2

Machine Learning-Driven Approaches for Dermatological Diseas...

引用

2nd IEEE International Conference on Data science and Network Security, ICDSNS 2024

作者： Raman, Ramakrishnan Kumar, Vikram Saini, Dimple Rabadiya, Dhaval Patre, Smruti Meenakshi, R. India Srv Media Sri Balaji University Pune India Parul Institute of Technology Parul University Computer Science and Engineering Engineering and Technology India Pune India Chennai Institute of Technology Deparment of Computer Science and Engineering India

ISBN: (纸本)9798350373110

Recent advancements in machine learning (ML) have significantly enhanced the accuracy and efficiency of diagnosing dermatological diseases, surpassing conventional methods. This paper presents a detailed exploration of ML-driven approaches, employing sophisticated algorithms such as Convolutional Neural Networks (CNNs), Support Vector Machines (SVMs), and Decision Trees to develop a predictive model for various skin conditions. Our methodology involved collecting and preprocessing a comprehensive dataset of dermatological images, followed by applying these algorithms to classify and predict skin diseases. The model was rigorously evaluated against traditional diagnostic methods using metrics such as accuracy, precision, recall, and F1 score. Results demonstrate that our ML-driven model, particularly the CNNs, achieved superior diagnostic performance with an accuracy of 94%, precision of 92%, recall of 93%, and an F1 score of 92%. These findings highlight the potential of ML to revolutionize dermatological diagnostics, offering a more accessible, efficient, and precise approach to identifying skin diseases. The study underscores the importance of integrating artificial intelligence in healthcare, paving the way for advancements in dermatological diagnostic tools and contributing to the broader discussion on AI in medical applications. This work not only enhances diagnostic capabilities but also opens avenues for more equitable healthcare solutions by reducing reliance on limited dermatological expertise. © 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

Harnessing AI for Cybersecurity: Performance Evaluation of Machine Learning Techniques in Intrusion Detection Systems 1

Harnessing AI for Cybersecurity: Performance Evaluation of M...

引用

1st International Conference on Advances in computer science, Electrical, Electronics, and Communication Technologies, CE2CT 2025

作者： Kumar, Naween Sharma, Supragya Gupta, Sahil Jain, Sajal Singh, Sujal School of Computer Science Engineering & Technology Bennett University Greater Noida India

ISBN: (纸本)9798331518578

Due to the current development of networked systems and the emergence of global cyber threats, efficient and fast NIDS are required. The suggested article aims to analyse new and improved AI and Machine Learning (ML) frameworks to design and implement efficient Network Intrusion Detection Systems (NIDSs) that accurately detect and differentiate network intrusions. The article assesses several effective algorithms for intrinsic and practical use: Logistic Regression, Random Forest, Feedforward Neural Networks (FNN), K Nearest Neighbours (KNN), and Long Short-Term Memory (LSTM), in terms of actual positive intrusions without creating high positive false intrusions. A comparative analysis is then made to evaluate the models based on precision, recall rate, accuracy, and the model processing rate. This suggested article presents a comparison of the two techniques regarding their effectiveness and applicability of the techniques for implementation in the next generation of NIDS. Although observations showed good performance and improvement regarding class imbalance, they remain key issues affecting such systems' reliability, including high computational demand and model interpretability. The reported results such as accuracy 99.897%, precision 99.894%, recall 99.898%, and F1 Score 99.892% in case of Logistic Regression and accuracy 99.882%, precision 99.876%, recall 99.882%, and F1 Score 99.892% in case of Random Forest from the work contribute to the advancement of fully intelligent NIDS that will improve the system stability to defend and adapting for new emerging threats, hence improving the advance of technologies in the complex and sophisticated era of cyberstalking. © 2025 IEEE.

关键词： Network intrusion

来源：评论

学校读者我要写书评

暂无评论

Performance and pollutant characteristics of an old generation diesel engine fuelled with dual-fuel diesel-dimethyl ether

引用

International Journal of Ambient Energy 2023年第1期44卷 555-563页

作者： Cong, Doan Nguyen Duc, Khanh Nguyen Duy, Vinh Nguyen Faculty of Mechanical Engineering University of Transport Technology Hanoi Viet Nam Department of Vehicle and Energy Conversion Engineering School of Mechanical Engineering Hanoi University of Science and Technology Hanoi Viet Nam Faculty of Vehicle and Energy Engineering Phenikaa University Hanoi Viet Nam

An old generation diesel engine running on dual-fuel dimethyl ether (DME) and diesel was tested for pollution reduction and performance enhancement. DME was pre-mixed with intake air using a single-port gaseous fuel injection system to replace 5% (DME5) and 10% (DME10) diesel. Study results show that the fuel economy improves when the engine operates with dual fuel. The brake-specific energy consumption reduces by 2.5% and 3.0% at full load condition and 4.5% and 5.1% at a medium speed of 1600 rpm on average for DME5 and DME10 compared to pure diesel case. The in-cylinder pressure, temperature, and vibration rate were evaluated to determine DME's effect on engine combustion. The impact of DME was noticeable as exhaust pollutants enhanced remarkably for NOx and smoke components. When the engine was fueled with DME5 and DME10, the specific particulate matter fell dramatically by 32.90% and 50.23%, NOx declined moderately by 8.66% and 20.39%. © 2022 Informa UK Limited, trading as Taylor & Francis Group.

关键词： Diesel engines

来源：评论

学校读者我要写书评

暂无评论

A survey on textual emotion cause extraction in social networks

引用

Digital Communications and Networks 2025年第2期11卷 524-536页

作者： Sancheng Peng Lihong Cao Guojun Wang Zhouhao Ouyang Yongmei Zhou Shui Yu Laboratory of Language Engineering and Computing Center for Linguistics and Applied LinguisticsGuangdong University of Foreign StudiesGuangzhou510006China School of English Education Guangdong University of Foreign StudiesGuangzhou510006China School of Computer Science and Cyber Engineering Guangzhou UniversityGuangzhou510006China South China University of Technology Guangzhou510006China School of Information Science and Technology Guangdong University of Foreign StudiesGuangzhou510006China School of Computer Science University of Technology SydneySydneyNSW 2007Australia

With the rapid development of web technology,Social Networks(SNs)have become one of the most popular platforms for users to exchange views and to express their *** and more people are used to commenting on a certain hot spot in SNs,resulting in a large amount of texts containing *** Emotion Cause Extraction(TECE)aims to automatically extract causes for a certain emotion in texts,which is an important research issue in natural language *** is different from the previous tasks of emotion recognition and emotion *** addition,it is not limited to the shallow-level emotion classification of text,but to trace the emotion *** this paper,we provide a survey for ***,we introduce the development process and classification of ***,we discuss the existing methods and key factors for ***,we enumerate the challenges and developing trend for TECE.

关键词： Text Emotion Emotion cause Machine learning Deep learning

来源：评论

学校读者我要写书评

暂无评论

FCMDNet:A Feature Correlation-Based Multi-Scale Dual-Branch Network for Infrared and Visible Image Fusion 6

FCMDNet:A Feature Correlation-Based Multi-Scale Dual-Branch ...

引用

6th International Conference on Robotics and computer Vision, ICRCV 2024

作者： Jiang, Zheng Chen, Xiangyang Zan, Guo Yang, Mingyu School of Computer Science and Engineering Wuhan Institute of Technology Wuhan China

ISBN: (纸本)9798331527426

In recent years, deep learning techniques have been widely used in the field of infrared and visible image fusion (IVIF) due to their powerful feature representation. Existing methods usually use convolutional neural network (CNN), Transformer or Transformer-CNN to process different modal images. However, if the relationship between modality-specific and modality-shared features is ignored in the fusion network, the fused image can result in disturbed or missing modality information. To solve this problem, we propose a multi-scale dual-branch fusion network based on Transformer and CNN, called FCMDNet. Firstly, FCMDNet network introduces Multi-scale Patch Embedding based on Restormer module to extract multi-scale shallow features across modalities. Then, we design a dual-branch Transformer-CNN feature extractor, where the Transformer branch exploits the long and short-range attention of the Lite Transformer for low-frequency global features, the CNN branch uses residual dense connections to extract high-frequency local features. In addition, we propose a loss function based on feature correlation that exploits the correlation of low-frequency features and the uncorrelation of high-frequency features between different modalities to improve the fusion quality. Finally, an addition fusion strategy and two fusion layers are used to fuse the global and local features to obtain the final fused image. Experimental results on two public datasets show that our proposed FCMDNet has good performance on all of them. © 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

Remember This Event That Year? Assessing Temporal Information and Understanding in Large Language Models

Remember This Event That Year? Assessing Temporal Informatio...

引用

2024 Conference on Empirical Methods in Natural Language Processing, EMNLP 2024

作者： Beniwal, Himanshu Patel, Dishant Kowsik Nandagopan, D. Ladia, Hritik Yadav, Ankit Singh, Mayank Department of Computer Science and Engineering Indian Institute of Technology Gandhinagar India

ISBN: (纸本)9798891761681

Large Language Models (LLMs) are increasingly ubiquitous, yet their ability to retain and reason about temporal information remains limited, hindering their application in real-world scenarios where understanding the sequential nature of events is crucial. Our study experiments with 12 state-of-the-art models (ranging from 2B to 70B+ parameters) on a novel numerical-temporal dataset, TempUN, spanning from 10,000 BCE to 2100 CE, to uncover significant temporal retention and comprehension limitations. We propose six metrics to assess three learning paradigms to enhance temporal knowledge acquisition. Our findings reveal that open-source models exhibit knowledge gaps more frequently, suggesting a trade-off between limited knowledge and incorrect responses. Additionally, various fine-tuning approaches significantly improved performance, reducing incorrect outputs and impacting the identification of 'information not available' in the generations. The associated dataset and code are available at https://***/lingoiitgn/TempUN. © 2024 Association for Computational Linguistics.

关键词： Computational linguistics

来源：评论

学校读者我要写书评

暂无评论

Cross-lingual Editing in Multilingual Language Models 18

Cross-lingual Editing in Multilingual Language Models

引用

18th Conference of the European Chapter of the Association for Computational Linguistics, EACL 2024 - Findings of EACL 2024

作者： Beniwal, Himanshu Kowsik Nandagopan, D. Singh, Mayank Department of Computer Science and Engineering Indian Institute of Technology Gandhinagar India

ISBN: (纸本)9798891760936

The training of large language models (LLMs) necessitates substantial data and computational resources, and updating outdated LLMs entails significant efforts and resources. While numerous model editing techniques (METs) have emerged to efficiently update model outputs without retraining, their effectiveness in multilingual LLMs, where knowledge is stored in diverse languages, remains an underexplored research area. This research paper introduces the cross-lingual model editing (XME) paradigm, wherein a fact is edited in one language, and the subsequent update propagation is observed across other languages. To investigate the XME paradigm, we conducted experiments using BLOOM, mBERT, and XLM-RoBERTa using the two writing scripts: Latin (English, French, and Spanish) and Indic (Hindi, Gujarati, and Bengali). The results reveal notable performance limitations of state-of-the-art METs under the XME setting, mainly when the languages involved belong to two distinct script families. These findings highlight the need for further research and development of XME techniques to address these challenges. For more comprehensive information, the dataset used in this research and the associated code are publicly available at the following URL. © 2024 Association for Computational Linguistics.

关键词： Computational linguistics

来源：评论

学校读者我要写书评

暂无评论

DeepFrame: A Scientific Framework for Advancing Ocean Cleanup through Underwater Debris Detection and Geospatial Mapping

DeepFrame: A Scientific Framework for Advancing Ocean Cleanu...

引用

2024 International Conference on Innovative Computing, Intelligent Communication and Smart Electrical Systems, ICSES 2024

作者： Jesher Joshua, M. Ragav, V. Pandiyaraju, V. School of Computer Science and Engineering Vellore Institute of Technology Chennai India

ISBN: (数字)9798331543624

ISBN: (纸本)9798331543617

The escalating concern of underwater debris demands innovative approaches for effective ocean cleanup. This study presents a comprehensive framework utilizing deep learning methods to detect, categorize, track, and geospatially map un derwater debris. By amalgamating cutting-edge object detection models, the system extracts real-time insights from underwater sensor data, facilitating the identification and categorization of debris based on various attributes. The incorporation of advanced tracking algorithms enables the monitoring of debris dynamics, contributing to predictive capabilities. Furthermore, geospatial mapping techniques are employed to project identified debris onto world maps, using latitude and longitude coordinates. This mapping feature aids in locating debris accumulation areas, thereby enabling strategic and efficient cleanup efforts. Through meticulous experimentation on authentic underwater debris datasets, our framework demonstrates its potential in precisely detecting, tracking, and mapping debris. The proposed solution leverages robust extended dataset of augmented images with its benchmark and novel Geo-Spatial mapping framework, which revolutionizes ocean conservation endeavors by providing valuable insights and ecologically-aware of underwater debris removal. The source code and the dataset have been made publicly available on GitHubl. © 2024 IEEE.

关键词： Mapping

来源：评论

学校读者我要写书评

暂无评论

Using Blended Predictions made through using Unlabeled Data and Fast Text computers from Internet Companies, Unhappiness May be Detected 11

Using Blended Predictions made through using Unlabeled Data ...

引用

11th International Conference on System Modeling and Advancement in Research Trends, SMART 2022

作者： Kumar Sinha, Deepak Shreenidhi, H.S. Faculty of Engineering and Technology Department of Computer Science and Engineering Bangalore India

ISBN: (纸本)9781665487320

Disease is a significant contributor to both suicides and or the biggest cause ill long - term disability. Vernacular usage is significantly impacted by sorrow. In this research, we use a number of classifiers that are focused on communications and postings on social networking sites to deliver in early detection of anxiety. Among the supervised classification strategies are Matrix, SVM, Unusual, and Arched. The classifier employing phonetic characteristics at the point - of - care has been linked to the actress's training to use the Classifier, Faster Text deep feature approach. © 2022 IEEE.

关键词： Support vector machines

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：