检索结果-内蒙古大学图书馆

2024 International Conference on Innovation and Intelligence for Informatics, Computing, and Technologies, 3ICT 2024

作者： Rane, Rashmi Subhashini, R. Surendran, R. Department of Computer Science and Engineering School of Computing Sathyabama Institute of Science and Technology Chennai India School of Computer Engineering and Technology Dr. Vishwanath Karad World Peace University Pune India Department of Computer Science and Engineering SRM Institute of Science and Technology Ramapuram Chennai India Department of Computer Science and Engineering Saveetha School of Engineering Saveetha Institute of Medical and Technical Sciences Chennai India

ISBN: (数字)9798331533137

ISBN: (纸本)9798331533137

Rapid developments in digital technology have expedited the dissemination of information on social media platforms like as Twitter, Facebook, and Weibo. Unverified information can create protests and mislead the public. It can also cause significant issues for governmental entities and society at large. Unverified content is frequently shared as text, images, audio, or video. This paper introduces a novel Self-Guided Edge Attention-Focused Transformer Slimmable Network improved by Osprey Optimization (SGEAFTSN-OO) for the purpose of identifying fake news. This system operates using the Politifact and Gossipcop databases. Text and visual data are preprocessed using NLP Stemming Self-Guided Filtering (NLPS-SGF). To extract features, the Multilayer Edge Attention with Focused Transformer (MEAFT) is then applied. Lastly, Slimmable Pruned Neural Network with Osprey Optimization (SP-Net-OO) algorithm is employed to identify fake news events. Metrics including accuracy, recall, precision, and F1-score are compared with those of the existing approaches in order to assess the effectiveness of the suggested strategy. The proposed model achieved 99.21% accuracy, 99.15% precision, and 99.12% F1-score based on experimental data. ©2024 IEEE.

关键词： Tweets

来源：评论

学校读者我要写书评

暂无评论

RobustFace: a novel image restoration technique for face adversarial robustness improvement

引用

Multimedia Tools and Applications 2025年第11期84卷 8965-8993页

作者： Sadu, Chiranjeevi Das, Pradip K. Yannam, V Ramanjaneyulu Nayyar, Anand Department of Computer Science and Engineering RGUKT Nuzvid Andhra Pradesh Nuzvid India Department of Computer Science and Engineering IIT Guwahati Assam Guwahati India School of Technology Woxsen University Telangana Hyderabad India Graduate School Faculty of Information Technology Duy Tan University Da Nang550000 Viet Nam

Machine Learning (ML) models, particularly Deep Learning (DL), have made rapid progress and achieved significant milestones across various applications, including numerous safety-critical contexts. However, these models have recently been discovered to be susceptible to adversarial attacks, which are well-crafted input images. The primary objective of this paper is to propose a novel methodology titled "RobustFace", which is based on deep image restoration networks, that significantly improves the facial adversarial robustness of various image-classification models. Adversarial images are created using the Private Fast Gradient Sign Method (P-FGSM), StyleGAN and Fast Landmark Manipulation (FLM) methods. The adversarial images are then enhanced using deep image restoration networks to bring back them into the original space. The encoded weighted local magnitude patterns (WLMP) are extracted and provided to different types of classifiers to detect facial adversarial images from the clean images. The effectiveness of RobustFace has been demonstrated on two real-world datasets and experimental outcomes show that it significantly improves facial adversarial robustness on all evaluating classifiers. It improves the highest classification accuracy from 98.75% to 99.00% on P-FGSM attacks, from 77.94% to 85.25% on adversarial attacks generated by StyleGAN and from 65.52% to 69.50% for FLM attacks. © The Author(s), under exclusive licence to Springer science+Business Media, LLC, part of Springer Nature 2024.

关键词： Image reconstruction

来源：评论

学校读者我要写书评

暂无评论

Learning Temporal User Features for Repost Prediction with Large Language Models

引用

computers, Materials & Continua 2025年第3期82卷 4117-4136页

作者： Wu-Jiu Sun Xiao Fan Liu School of Computer Science and Engineering Southeast UniversityNanjing11189China Department of Media and Communication City University of Hong KongHong Kong999077 China

Predicting information dissemination on social media,specifcally users’reposting behavior,is crucial for applications such as advertising *** methods use deep neural networks to make predictions based on features related to user topic interests and social ***,these models frequently fail to account for the difculties arising from limited training data and model size,which restrict their capacity to learn and capture the intricate patterns within microblogging *** overcome this limitation,we introduce a novel model Adapt pre-trained Large Language model for Reposting Prediction(ALL-RP),which incorporates two key steps:(1)extracting features from post content and social interactions using a large language model with extensive parameters and trained on a vast corpus,and(2)performing semantic and temporal adaptation to transfer the large language model’s knowledge of natural language,vision,and graph structures to reposting prediction ***,the temporal adapter in the ALL-RP model captures multi-dimensional temporal information from evolving patterns of user topic interests and social preferences,thereby providing a more realistic refection of user ***,to enhance the robustness of feature modeling,we introduce a variant of the temporal adapter that implements multiple temporal adaptations in parallel while maintaining structural *** results on real-world datasets demonstrate that the ALL-RP model surpasses state-of-the-art models in predicting both individual user reposting behavior and group sharing behavior,with performance gains of 2.81%and 4.29%,respectively.

关键词： Reposting prediction large language model semantic adaptation temporal adaptation

来源：评论

学校读者我要写书评

暂无评论

WSETO: wild stock exchange trading optimization algorithm enabled routing for NB-IoT tracking system

引用

International Journal of Information Technology (Singapore) 2024年第8期16卷 4895-4905页

作者： Sivadevuni, Sreeparnesh Sharma Naveen, J. Department of Computer Science and Engineering School of Engineering and Technology Christ University Karnataka Bangalore 560001 India

The Narrowband Internet of Things (NB-IoT) communication plays a significant role in the IoT due to the capability of generating broad exploration with the usage of limited power. Over the past few years, the Low Power Wide Area Networks (LPWAN) have been efficient in the data acquisition and remote monitoring area however they failed to generate high data rates, low latency, and the consumption of low power. To solve these problems, NB-IoT technology has developed in long-term asset tracking and it replaces the Global Positioning System (GPS) with its ubiquitous coverage. In this research, the Wild Stock Exchange Trading Optimization technique (WSETO) is proposed for a routing-based NB-IoT tracking system. The WSETO is the combination of the Wild Geese Algorithm (WGA) and SETO. By employing WSETO, the routing to the relevant target location is established effectively. The existing techniques like Low Power Asset Tracking of NB-IoT (LoPATraN), Monitoring system based on NB-IoT and BeiDou System/GPS (BDS/GPS), and Narrowband Physical Uplink Shared Channel (NPUSCH) are used to compare the WSETO approach. In rounds with a value of 2000, the WSETO demonstrates a superior location error of 0.001 in comparison to existing methods such as LoPATraN, a monitoring system utilizing NB-IoT and BDS/GPS, as well as NPUSCH. © Bharati Vidyapeeth's Institute of computer Applications and Management 2024.

关键词： Global positioning system Low power wide area networks Narrowband internet of things Stock exchange trading optimization Wild geese algorithm

来源：评论

学校读者我要写书评

暂无评论

Stacked LSTM for Solar Irradiance Forecasting Using Minimum Redundancy Maximum Relevance (MRMR) Feature Selection 1

Stacked LSTM for Solar Irradiance Forecasting Using Minimum ...

引用

1st International Conference on Trends in engineering Systems and Technologies, ICTEST 2024

作者： Hari, Ninu G. Jisha, G. Rajagiri School of Engineering and Technology Apj Abdul Kalam Technological University Department of Computer Science and Engineering Kerala India Rajagiri School of Engineering and Technology Department of Computer Science and Engineering Kochi India

ISBN: (纸本)9798350307795

To effectively combat atmospheric pollution caused by greenhouse gases, immediately switching to power plants that rely solely on renewable energy sources is imperative. With the vast availability of solar energy in Kerala throughout the year, it would be wise to make it the primary source for these plants. The spatiotemporal features that impact the Global Horizontal Irradiance (GHI) should also be analysed to choose a better location for solar power plants. The features should be identified and analysed yearly and seasonally to determine the relationship between GHI and those selected features. The Minimum Redundancy Maximum Relevance (MRMR) minimal optimal feature selection algorithm is integrated for seasonal and yearly GHI prediction for finding the feature relation. The stacked LSTM (Long Short-Term Memory) with a 3-layer architecture is used for solar irradiance forecasting and annual and seasonal analysis of the dataset since LSTM is a perfect model for time-series data and stacked LSTM performs better than single-layer LSTM. The study shows that feature identification and selection positively change the forecasting results. The comparison of results also shows that seasonal forecasting needs more focus than yearly forecasting since results fluctuate in different seasons. The tropical monsoon climate forecasting is more complex as it fluctuates most of the year and is considered for experimental analysis. © 2024 IEEE.

关键词： Forecasting

来源：评论

学校读者我要写书评

暂无评论

Scalable alignment-free feature extraction approach for genome data and their cluster analysis

引用

Multimedia Tools and Applications 2025年 1-37页

作者： Tripathi, Abhishek Tiwari, Aruna Chaudhari, Narendra S. Ratnaparkhe, Milind Bharill, Neha Jha, Preeti Dwivedi, Rajesh Department of Computer Science IIT Indore Simrol Madhya Pradesh Indore453552 India ICAR-Indian Institute of Soybean Research Indore Madhya Pradesh Indore452001 India Department of Computer Science Mahindra University Ecole Centrale School of Engineering Telangana Hyderabad500043 India Department of Computer Science and Engineering Koneru Lakshmaiah Education Foundation Bowrampet Telangana Hyderabad500043 India

Feature extraction is crucial in bioinformatics, as it converts genomic sequences into numerical feature vectors essential for machine learning algorithms, particularly in clustering, to identify the families of newly sequenced genomes. Traditional methods have relied on alignment-based techniques for clustering the genomic sequences. However, these methods are computationally intensive. In contrast, alignment-free methods are now more commonly used due to their reduced computational demands. Despite this, many alignment-free approaches may generate identical feature vectors for dissimilar sequences, as they focus solely on single nucleotide counts (1-gram) and their arrangement during feature extraction, often neglecting dinucleotide counts and their arrangement, which can degrade clustering performance. Furthermore, certain approaches include trinucleotide or higher-order compositions;they introduce high-dimensionality issues, resulting in inaccurate results. Additionally, some existing methods are not scalable and take substantial time to extract features from large genomic sequences. To address these issues, we proposed a novel 33-dimensional Scalable Alignment-Free Feature Vector (33d-SAFFV) approach to extract the significantly important features such as length of sequence, count of dinucleotides, and positional sum of dinucleotides, which produces a 33-dimensional feature vector. This approach leverages Apache Spark for scalability and efficient in-memory computations, making it suitable for large datasets. We evaluated the performance of our proposed method by applying the extracted 33-dimensional feature vectors to K-Means and Fuzzy C-Means (FCM) clustering algorithms. Performance is measured using the Silhouette Index (SI) and Calinski-Harabasz (CH) index. Experimental results on the gene sequences of four varieties of rice datasets and two varieties of soybean datasets show the effectiveness of the proposed 33d-SAFFV approach. In K-Means clustering with t

关键词： Nucleotides

来源：评论

学校读者我要写书评

暂无评论

Occluded human pose estimation based on limb joint augmentation

引用

Neural Computing and Applications 2025年第3期37卷 1241-1253页

作者： Han, Gangtao Song, Chunxiao Wang, Song Wang, Hao Chen, Enqing Wang, Guanghui School of Electrical and Information Engineering Zhengzhou University Zhengzhou China School of Computer and Artificial Intelligence Zhengzhou University Zhengzhou China Department of Computer Science Toronto Metropolitan University Toronto Canada

Human pose estimation aims at locating the specific joints of humans from the images or videos. While existing deep learning-based methods have achieved high positioning accuracy, they often struggle with generalization in occlusion scenarios. In this paper, we propose an occluded human pose estimation framework based on limb joint augmentation to enhance the generalization ability of the pose estimation model on the occluded human bodies. Specifically, the occlusion blocks are at first employed to randomly cover the limb joints of the human bodies from the training images, imitating the scene where the objects or other people partially occlude the human body. Trained by the augmented samples, the pose estimation model is encouraged to accurately locate the occluded keypoints based on the visible ones. To further enhance the localization ability of the model, this paper constructs a dynamic structure loss function based on limb graphs to explore the distribution of occluded joints by evaluating the dependence between adjacent joints. Extensive experimental evaluations on two occluded datasets, OCHuman and CrowdPose, demonstrate significant performance improvements without additional computation cost during inference. © The Author(s), under exclusive licence to Springer-Verlag London Ltd., part of Springer Nature 2024.

关键词： Joints (anatomy)

来源：评论

学校读者我要写书评

暂无评论

Enhanced Cyclone Intensity Estimation Through CNN Analysis of INSAT-3D IR Imagery 12th

Enhanced Cyclone Intensity Estimation Through CNN Analysis o...

引用

12th International Conference on Recent Trends in Computing, ICRTC 2024

作者： Kumar, Divvela Vishnu Sai Arora, Deepak Tiwari, Shivam Department of Computer Science and Engineering Amity School of Engineering and Technology Amity University Lucknow Campus Lucknow India Department of Computer Science and Engineering Sharda University Greater Noida India

ISBN: (纸本)9789819788354

Cyclone forecasting using satellite pictures involves anticipating the cyclone’s intensity in advance of its arrival. The results of this study can inform people’s preparations for the cyclone. In order to save lives and reduce damage, cyclone strength forecast is crucial. As a result of the many dangers they pose, these cyclones are among the most expensive natural disasters that have ever occurred. Five separate CNN (Convolutional Neural Network) models are presented by the authors to assess the cyclone’s severity using INSAT-3D infrared pictures. While many methods exist for making such an assessment, the authors’ top model yields an RMSE of 10.02 kts for intensity. © The Author(s), under exclusive license to Springer Nature Singapore Pte Ltd. 2025.

关键词： Convolutional neural networks

来源：评论

学校读者我要写书评

暂无评论

Multi-class Chest Radiograph Classification Using Deep Convolutional Generative Adversarial Networks 6th

Multi-class Chest Radiograph Classification Using Deep Convo...

引用

6th International Conference on Recent Innovations in Computing, ICRIC 2023

作者： Kaushik, Hariharan Natarajan, B. Annamalai, R. Department of Computer Science and Engineering Amrita School of Computing Amrita Vishwa Vidyapeetham Chennai India School of Computer Science and Engineering VIT University Chennai India

ISBN: (纸本)9789819778614

The urgency for precise diagnostics during the COVID-19 pandemic has driven advancements in imaging and deep learning tools. However, progress is impeded by limited access to medical imaging data. This study employs cutting-edge deep learning techniques to identify chest diseases, including COVID-19 and various lung conditions. State-of-the-art Generative Adversarial Networks (GANs), including CycleGAN, Super-Resolution GAN (SRGAN), and Deep Convolutional GAN (DCGAN), are used to synthesize medical images, augmenting the dataset and enhancing model robustness. VGG-19 achieves a remarkable 91% accuracy and an F1-score of 0.98 across diverse chest-related diseases. Saliency Maps and Grad-CAM graphics improve interpretability, showcasing the model’s healthcare potential. The study encompasses diseases such as Pneumonia and COVID-19, contributing to a comprehensive classification task. These findings underscore the impact of deep learning in medical imaging, promising improved patient care and healthcare accessibility. © The Author(s), under exclusive license to Springer Nature Singapore Pte Ltd. 2024.

关键词： Generative adversarial networks

来源：评论

学校读者我要写书评

暂无评论

SACNN-IDS: A self-attention convolutional neural network for intrusion detection in industrial internet of things

引用

CAAI Transactions on Intelligence Technology 2024年第6期9卷 1398-1411页

作者： Mimonah Al Qathrady Safi Ullah Mohammed S.Alshehri Jawad Ahmad Sultan Almakdi Samar M.Alqhtani Muazzam A.Khan Baraq Ghaleb Department of Information Systems College of Computer Science and Information SystemsNajran UniversityNajranSaudi Arabia Department of Computer Science Quaid-i-Azam UniversityIslamabadPakistan Department of Computer Science College of Computer Science and Information SystemsNajran UniversityNajranSaudi Arabia School of Computing Engineering and the Built EnvironmentEdinburgh Napier UniversityEdinburghUK ICESCO Chair Big Data Analytics and Edge Computing Quaid-i-Azam UniversityIslamabadPakistan

Industrial Internet of Things(IIoT)is a pervasive network of interlinked smart devices that provide a variety of intelligent computing services in industrial *** IIoT nodes operate confidential data(such as medical,transportation,military,etc.)which are reachable targets for hostile intruders due to their openness and varied *** Detection Systems(IDS)based on Machine Learning(ML)and Deep Learning(DL)techniques have got significant ***,existing ML and DL-based IDS still face a number of obstacles that must be *** instance,the existing DL approaches necessitate a substantial quantity of data for effective performance,which is not feasible to run on low-power and low-memory *** and fewer data potentially lead to low performance on existing *** paper proposes a self-attention convolutional neural network(SACNN)architecture for the detection of malicious activity in IIoT networks and an appropriate feature extraction method to extract the most significant *** proposed architecture has a self-attention layer to calculate the input attention and convolutional neural network(CNN)layers to process the assigned attention features for *** performance evaluation of the proposed SACNN architecture has been done with the Edge-IIoTset and X-IIoTID *** datasets encompassed the behaviours of contemporary IIoT communication protocols,the operations of state-of-the-art devices,various attack types,and diverse attack scenarios.

关键词： convolutional neural network deep learning industrial internet of things intrusion detection self-attention

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：