检索结果-内蒙古大学图书馆

International Journal of computers and Applications 2023年第5期45卷 379-390页

作者： Liu, J. Shao, H.J. Deng, X. Jiang, Y.T. School of Computer Jiangsu University of Science and Technology Zhenjiang China Department of Electrical and Computer Engineering University of Nevada Las VegasNV United States

The paper discusses the high computational costs associated with convolutional neural networks (CNNs) in real-world applications due to their complex structure, primarily in hidden layers. To overcome this issue, the paper proposes a novel channel pruning technique that leverages the correlation topology of feature maps generated by each CNNs layer to construct a network with fewer nodes, reducing computational costs significantly. Redundant channels exhibit a high degree of topological similarity and tend to increase as the number of network layers rises. Removing the channel corresponding to highly correlated feature maps allows retrieval of the ‘base’ set of characteristics needed by subsequent layers. The proposed channel pruning technique provides a promising approach to reducing the computational costs of deep convolutional neural networks while maintaining high performance levels. By designing a network structure optimized for specific input data types, the method results in more efficient and effective machine learning models. The pruning operation requires fine-tuning to optimize network performance, and experiments using X-ray, chest CT, and MNIST images show that the pruned network can eliminate approximately 80% of redundant channels with minimal performance deterioration (maintaining original CNNs performance at 99.2%). © 2023 Informa UK Limited, trading as Taylor & Francis Group.

关键词： Topology

来源：评论

学校读者我要写书评

暂无评论

Fake News Detection: Exploring the Efficiency of Soft and Hard Voting Ensemble 4

Fake News Detection: Exploring the Efficiency of Soft and Ha...

引用

4th International Conference on Evolutionary Computing and Mobile Sustainable Networks, ICECMSN 2024

作者： Rahman, Arifur Zaman, Sakib Parvej, Shahriar Shill, Pintu Chandra Salim, Md. Shahidul Das, Dola Department of Computer Science and Engineering Khulna University of Engineering & Technology Khulna9203 Bangladesh

Fake news dissemination is a critical problem, jeopardizing the reliability of information and eroding public confidence. Machine learning provides a solution by distinguishing between real and misleading news. In this study, we utilized various classifiers such as KNN, Logistic Regression, Decision Tree, LightGBM, XGBoost, Gradient Boosting, Random Forest, and Naive Bayes on an open-source dataset. Tf-idf (Term frequency-inverse document frequency) and Count vectorization were used for text preprocessing, and soft and hard voting ensembles were applied to the best three and five models to boost performance. Additionally, we employed BERT (Bidirectional Encoder Representations from Transformers) and CNN (Convolutional Neural Network) models with various optimizers (adam, rmsprop, SGD, adadelta, and adagrad) to further improve classification. The BERT model, optimized with adam, achieved the highest accuracy of 99.93%, with perfect precision (100%) and high recall (99.84%). Furthermore, using Count vectorization for feature extraction, the soft voting ensemble of the top five models (LGBM, XGB, RF, DT, LR) achieved peak performance among all ensemble models, achieving 99.87% accuracy, with precision, recall, f1-score, MAE, MSE, RMSE, RAE, and RRSE of 99.83%, 99.88%, 99.86%, 0.13%, 0.13%, 3.66%, 0.27%, and 7.80%, respectively. The exhaustive experimentation confirms the applicability and efficiency of the recommended models in identifying fake news. © 2025 The Authors. Published by Elsevier B.V.

关键词： Convolutional neural networks

来源：评论

学校读者我要写书评

暂无评论

Exploring the potential of cloud computing in revolutionizing healthcare 4th

Exploring the potential of cloud computing in revolutionizin...

引用

4th International Conference on Computational Methods in science and Technology, ICCMST 2024

作者： Saini, Renu Paras Tayal, Parag Thakur, Pallavi Computer Science and Engineering Department Chandigarh Engineering College Mohali Punjab Landran India

ISBN: (纸本)9781032911571

Cloud computing, which moves assets from desktop setting to cloud infra-structure—where apps are kept in a "cloud" with massive processing power and nearly infinite database storage—has fundamentally altered traditional computing over the past few years. Modern healthcare can benefit greatly from telemedicine when it is combined with cloud computing. Strategic thinking is necessary before moving healthcare assets and services to the cloud in order to weigh the advantages and disadvantages of the fresh approach, determine whether it may be implemented, and analyze its capabilities. The study illustrated how various technologies could be applied to big data analysis in healthcare using a cloud computing. In this research, we investigate the way cloud computing has changed the way in which healthcare organizations keep their data. © 2025 the Author(s).

关键词： Electronic health record

来源：评论

学校读者我要写书评

暂无评论

Optimizing Academic Notifications: An Attention-Centric Cloud-Based Approach 4

Optimizing Academic Notifications: An Attention-Centric Clou...

引用

4th International Conference on Ubiquitous Computing and Intelligent Information Systems, ICUIS 2024

作者： Thapa, Hritik Kansakar, Pranima Shrestha, Aron Koirala, Salina Kathmandu University Department of Computer Science and Engineering Dhulikhel Nepal

ISBN: (纸本)9798331529635

Effectively balancing user attention and delivering relevant information is key to a successful notification system. With students managing an ever-increasing array of study materials, deadlines, and updates, a unified notification system becomes crucial for consolidating information from various sources such as college websites, class moderators, and Learning Management Systems (LMS). This paper aims to design an attention-based notification system that enhances user experience by providing timely, relevant updates while minimizing information overload. Utilizing cloud computing and seamless integration with the Moodle platform, the system is expected to enhance notification efficiency and user engagement through a cohesive and centralized platform and ensure that students stay informed without being overwhelmed. The system proposes a cloud-based architecture to achieve scalability and security. At the same time, its focus on urgency-based notifications provides an innovative approach to obtaining the relevance and impact of each message, addressing key challenges in current academic notification practices. © 2024 IEEE.

关键词： Cloud platforms

来源：评论

学校读者我要写书评

暂无评论

An Experimental Study on Denoising the Images with Autoencoders 1

引用

6th International Conference on Communications and Cyber-Physical engineering, ICCCE 2023

作者： Venkata Abhishek, Gullipalli Sai Prathek, Kotha Durga Aryan, Kosuri Srinivas, Gorla Department of Computer Science and Engineering Gitam University Visakhapatnam India

ISBN: (数字)9789819971374

ISBN: (纸本)9789819971367

Noisy data is still one of the most common issues in modern data transmission. We can solve this by using deep learning with an autoencoder, a feature extraction method that can reduce noise. In the current study, we examine the ability of an Autoencoder to minimize noise on various structures and amounts of additive noise implemented with multiple optimizers. The experiment is run on a Convolutional Denoising Autoencoder (CDAE), primarily for image data. The experiment is run with additive noise with a noise factor of 0.03, such as Gaussian, Salt and Pepper, and Speckle. The proposed model outperformed image quality measures PSNR, MSE, SSIM, and RMSE. The reconstructed images after denoising are identical to the original image data. © The Author(s), under exclusive license to Springer Nature Singapore Pte Ltd. 2024.

关键词： Additive noise

来源：评论

学校读者我要写书评

暂无评论

Study and Analysis of Key-Predistribution Schemes Based on Hash Chain for WSN

Study and Analysis of Key-Predistribution Schemes Based on H...

引用

International Conference on Information Security, Privacy, and Digital Forensics, 2022

作者： Yadav, Kanhaiya Kumar Ahlawat, Priyanka Department of Computer Science and Engineering NIT Kurukshetra Kurukshetra India

ISBN: (纸本)9789819950904

In recent years, Wireless Sensor Network (WSN) has become a leading area of research. WSN is ad hoc, without infrastructure wireless network that connects a huge number of wireless sensors. WSN consists of multiple sensor nodes which have limited processing speed, storage capacity, communication bandwidth and base stations. They can be thought of as the network’s "sensing cells" and "brain", respectively. They are emerging as a better technology in the future due to their large range of applications in surveillance and people-related domains. This paper proposes a study and analysis of hash-based key predistribution for a better understanding of the uses of key predistribution or random key predistribution over the wireless sensor network to prevent the makers and also provide better connectivity, lower storage overhead, lower communication overhead and less computation complexities. © 2024, The Author(s), under exclusive license to Springer Nature Singapore Pte Ltd.

关键词： Sensor nodes

来源：评论

学校读者我要写书评

暂无评论

Rating-Based Sentiment Analysis on Online Mobile Reviews Using VADER Preprocessing

Rating-Based Sentiment Analysis on Online Mobile Reviews Usi...

引用

International Conference on Computing and Machine Learning, CML 2024

作者： Das, Tirtha Samanta, Subhrajit Basak, Piya Bhowmick, Ishita Daw, Swarnali Saha, Mousumi Computer Science and Engineering Narula Institute of Technology Kolkata India

ISBN: (纸本)9789819775484

Online shopping is rapidly increasing nowadays. As a reason when we are going to purchase any kind of item through an e-commerce site, we are always concerned about the reviews and ratings of the product that we are going to purchase. Naturally, review analysis has become very important in today’s era. The reviews are nothing but the opinions which are given by the user who purchased the item. So, opinion classification is very much trending. So, sentiment analysis from the opinion is a very much challenging issue in current days. Sentiment can be classified into different scales that can be good, bad, neutral, etc. and if we are also looking forward to the rating of the products, we are also looking forward to thoughts or judgment and likes or dislikes or other feelings and emotions are there. The dataset was collected from the Kaggle website based on a review of the mobile of Amazon (Rambocas and Gama in Marketing research: The role of sentiment analysis, Universidade do Porto, Faculdade de Economia do Porto, 2013) As Machine Learning (ML) is evolved, here we are going to use ML techniques such as ‘Naïve Bayes’, ‘Decision Tree’, ‘Random Forest’, ‘KNN’, and ‘Logistic Regression’ (Rambocas and Gama in Marketing research: The role of sentiment analysis, Universidade do Porto, Faculdade de Economia do Porto, 2013) for the rating and feature-based sentiment analysis on mobile reviews of Amazon website. Here we are going to use NLP and ML-based techniques to find out the review. So that users can easily buy the product from the analysis. The study verifies the effectiveness of the VADER tool by comparing it with manual tagging. The main contributions of this study are that it extracts the sentiment efficiently, achieves 92% accuracy with a decision tree, and gives practical insights to improve and understand public opinions. © The Author(s), under exclusive license to Springer Nature Singapore Pte Ltd. 2024.

关键词： Sentiment analysis

来源：评论

学校读者我要写书评

暂无评论

Animal Classification Using Cnn Architectures

Animal Classification Using Cnn Architectures

引用

2024 International Conference on Electrical, Electronics and Computing Technologies, ICEECT 2024

作者： Sharma, Dev Singh, Jayesh Isha Vipin Panwar, Divyanshi Akanksha, D. Anuj Lovely Professional University Computer Science and Engineering Phagwara India

ISBN: (纸本)9798350378092

This study addresses the problem of wild animal attacks and disappearance in areas such as farms or villages near forest and mountains, proposing a solution through deep learning algorithms. Using deep learning architectures, specifically convolution neural networks (CNNs), the study intends to accurately identify and classify animals based on categories. The dataset has been taken from Kaggle which consists of 90 different animal images, employing various CNN architectures those are LeNet-5, AlexNet, VGG16, ResNet50, Inception V3, and InceptionResnetV2. along with that implemented techniques such as data augmentation to fix the problem related to lack of data and, we used techniques like dropout to tackle the problem related to the model overfitting. The Study paper compares and analyses the accuracy of these algorithms, contributing to comprehensive understanding of their efficiency. Furthermore, different performance metrics are used names as accuracy, precision, recall and F1-score are used to check the accuracy and the model effectiveness. Furthermore, we have explored related works in automated classification and object detection, particularly focusing their application on animal classification. The results highlight the potential of deep learning in forest to create some sort of alert system or create a system which can provide information on endangered animal or rare animal. © 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

Deep Learning for Accurate Plant Disease Classification Using ResNet50: A Comprehensive Approach

Deep Learning for Accurate Plant Disease Classification Usin...

引用

2024 International Conference on Computational Intelligence and Computing Applications, ICCICA 2024

作者： Kumar, Manish Arora, Abhilaksh Deb, Arnab Yadav, Anup Lal Chandigarh University Department Of Computer Science And Engineering Punjab India

ISBN: (纸本)9798350306446

The timely and accurate identification of plant diseases is paramount for minimizing crop losses and guaranteeing food security globally. This study investigates the efficacy of deep learning in classifying plant diseases through the analysis of leaf images. A convolutional neural network (CNN) architecture is proposed, utilizing pre-trained weights from ResNet50 and subsequently fine-tuned on a dataset encompassing 38 distinct plant diseases. Data augmentation techniques are implemented to address class imbalance and bolster model robustness. The constructed model achieves a noteworthy validation accuracy of 96.49%, demonstrating its proficiency in differentiating various plant diseases. This research underscores the potential of deep learning in plant disease detection, paving the way for its real-world application within precision agriculture. © 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

Development and Evaluation of a Navigation System for Mobile Devices Using Vibration Positional Information 13

Development and Evaluation of a Navigation System for Mobile...

引用

13th IEEE Global Conference on Consumer Electronic, GCCE 2024

作者： Watanabe, Daichi Nakajima, Tatsuo Waseda University Department of Computer Science and Engineering Tokyo Japan

ISBN: (纸本)9798350355079

Currently, most mobile devices are equipped with vibration functions that notify users or simulate the feel of a touchscreen keyboard. However, most of these use a single motor for vibration, and the information about where the vibration occurs on the device is not included. To verify the usefulness of using vibration positional information for navigation, we have developed a navigation system that utilizes the position of vibrations. Our system provides tactile feedback through vibrations from motors placed at the four corners of a smartphone. We conducted a preliminary evaluation to verify the usefulness of using vibration position information for navigation. As a result, it was found that this system can provide appropriate and intuitive navigation, but issues related to intervals and patterns of the vibrations were pointed out. Finally, we discuss the future development of the system. © 2024 IEEE.

关键词： Smart devices

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：