检索结果-内蒙古大学图书馆

Lightweight Local-Global Fusion for Robust multiclass classification of Skin Lesions

INTERNATIONAL JOURNAL OF IMAGING SYSTEMS AND TECHNOLOGY 2025年第2期35卷

作者： Li, Guangli Zhou, Xinjiong Ye, Yiyuan Lv, Jingqin Ji, Donghong Wu, Jianguo Zhang, Ruiyang Zhang, Hongbin East China Jiaotong Univ Sch Informat & Software Engn Nanchang Peoples R China Wuhan Univ Sch Cyber Sci & Engn Wuhan Peoples R China Nanchang Univ Affiliated Hosp 2 Nanchang Peoples R China Commun Univ China Hainan Int Coll Beijing Peoples R China

Skin lesion classification is crucial for early diagnosis of skin cancer. However, the task faces challenges such as limited labeled data, data imbalance, and high intra-class variability. In this paper, we propose a lightweight local-global fusion (LGF) model that leverages the advantages of RegNet for local processing and Transformer for global interaction. The LGF model consists of four stages that integrate local and global pathological information using channel attention and residual connections. Furthermore, Polyloss is employed to address the data imbalance. Extensive experiments on the ISIC2018 and ISIC2019 datasets demonstrate that LGF achieves state-of-the-art performance with 93.10% and 90.36% accuracy, respectively, without any data augmentation. The LGF model is relatively lightweight and easier to reproduce, contributing to the field by offering a satisfactory trade-off between model complexity and classification performance. The code for our model will be available at .

关键词： dermoscopy images multiclass classification RegNet skin lesion classification transformer

来源：评论

学校读者我要写书评

暂无评论

Federated learning methods for collaborative multiclass classification of dry beans

引用

GENETIC RESOURCES AND CROP EVOLUTION 2025年第2期72卷 1421-1439页

作者： Gaur, Ankush Kumar Valan, J. Arul Natl Inst Technol Nagaland Dimapur 797103 India

Federated learning (FL), a decentralized approach, enables multiple institutions to collaboratively train a shared model while keeping their data localized. This ensures data privacy and security, crucial for sensitive agricultural datasets. This paper presents a comprehensive study on the application of FL methods for the collaborative multiclass classification of dry beans. We introduce the FL assisted agriculture 4.0 framework, which utilizes federated average and federated proximal aggregation techniques to create a consolidated model. Within this framework, a common multilayer perceptron model is trained to analyze decentralized agricultural data. For this study, we utilize the dry bean dataset from Kaggle for classification within a FL setup. The study evaluates the performance of the aforementioned FL algorithms in the context of multiclass classification, comparing them against traditional centralized models. To test the robustness of FL techniques on horizontally distributed agricultural data, we categorize the dataset into two types: independent and identically distributed (IID) and non-IID. In our simulation involving 10 clients, we explored scenarios featuring 0%, 40%, and 80% stragglers-nodes encountering delays. The results indicate that federated proximal outperformed federated average in handling delays and enhancing efficiency in distributed learning. Experimental results demonstrate that FL can achieve comparable, and in some cases superior, classification accuracy with significant improvements in data privacy. The implications of this study suggest a promising future for FL in agricultural applications, offering a pathway to more secure and collaborative machine learning frameworks.

关键词： Federated learning Federated optimization multiclass classification Collaborative learning

来源：评论

学校读者我要写书评

暂无评论

Two novel deep multi-view support vector machines for multiclass classification

引用

APPLIED INTELLIGENCE 2025年第2期55卷 1-17页

作者： Li, Yanfeng Xie, Xijiong Ningbo Univ Sch Informat Sci & Engn Ningbo 315211 Peoples R China

Multi-view classification methods have better generalization performance compared to the single-view classification methods due to the consistency information from multiple views. In recent years, the combination of support vector machine (SVM) and multi-view learning has been widely studied. To improve the robustness of multi-view classification methods, emphasis has shifted to the integration of multi-view classification approaches with fully-connected and convolutional neural networks. A classical deep two-view classification method named deep SVM-2K is a combination of support vector machine with two stage kernel canonical correlation analysis (SVM-2K) and deep learning. However, limitations of deep SVM-2K are that it can not cope with multi-view classification and multiclass classification problems. To address these issues, we propose two novel deep multi-view models named deep multi-view support vector machine (DMVSVM) for multiclass classification. DMVSVM uses the learned features by auto-encoder (AE) or deep neural network (DNN) to train the SVM classifier for each view. The two models then impose some constraints to make the output of the multi-view SVM classifiers as consistent as possible, which used to exploring intrinsic relations. Experiments performed on different real-word datasets show the effectiveness of our proposed approaches.

关键词： Deep multi-view learning SVM-2K multiclass classification Auto-encoder

来源：评论

学校读者我要写书评

暂无评论

An Effective Approach to Ensure Brand Safety in Online Advertising Using Image multiclass classification and Deep Learning 8th

An Effective Approach to Ensure Brand Safety in Online Adver...

引用

8th World Conference on Smart Trends in Systems Security and Sustainability, WorldS4 2024

作者： Cao, Nhan T. Vo, Quan M. Ton-That, An H. University of Information Technology Ho Chi Minh City Viet Nam Ho Chi Minh City Viet Nam Information Technology Faculty University of Finance and Marketing Ho Chi Minh City Viet Nam

ISBN: (纸本)9789819786947

Contextual advertising is a digital advertising method of displaying advertisements on the Internet according to the host website’s context. This leads to the need for businesses to ensure brand safety on the Internet, which is a strategy that enables a brand to avoid display of advertisement on illegal or inappropriate pages. This work proposes an effective approach to ensure brand safety online using multiclass classification to detect unsafe images for brands. We created a dataset with seven unsafe classes (adult, gore, gambling chip, gun, knife, alcohol, and cigarette) and a safe class. Moreover, some pruning techniques are investigated to observe the performance of our method. The performance is tested on certain key factors such as accuracy, model size, and inference speed. Experimental results demonstrate that our proposed approach obtains good performance and has potential to apply to real systems. © The Author(s), under exclusive license to Springer Nature Singapore Pte Ltd. 2025.

关键词： Brand safety Deep learning multiclass classification Online advertising Pruning techniques

来源：评论

学校读者我要写书评

暂无评论

Comparative Analysis of multiclass classification Using AE-CNN Method Combination on Nail Diseases

Comparative Analysis of Multiclass Classification Using AE-C...

引用

2025 International Conference on Advancement in Data Science, E-learning and Information System, ICADEIS 2025

作者： Palloge, Andi Herdiansyah Tahir, Zulkifli Syafaruddin Universitas Hasanuddin Makassar Department of Informatics Indonesia Universitas Hasanuddin Department of Electrical Makassar Indonesia

ISBN: (纸本)9798331513320

In the field of public health, there are various ways to diagnose multiple diseases in the human body. Diagnosing diseases at an early stage is essential. Multiple parts of the human body are observed to diagnose diseases, including the nails. The development of technology such as machine learning has brought many benefits in the health field, especially in disease detection, including the ability to detect nail diseases using nail images. This research aims to design and build a classification model for nail diseases such as Melanonychia, Nail dystrophy, normal, Onycholysis, and Onychomycosis. This research develops an AECNN model that combines an Autoencoder (AE) with a Convolutional Neural Network (CNN) with its architecture to perform multiclass classification of nail diseases. Each modeling in this study trains the model using images with noise to train the modeling on poor-quality photos. Then, the performance of the AE-CNN model was compared with several other models, namely AE-VGG16 and AE-ResNet50, based on accuracy, precision, recall, and F1-Score values. This study found that the proposed AE-CNN model was able to classify nail diseases better than the other models, with an accuracy of 91.8% for AE-CNN, 83.5% for AE-VGG16, and 48.3% for AE-ResNet50. © 2025 IEEE.

关键词： Autoencoder (AE) Convolutional Neural Network (CNN) multiclass classification nail diseases

来源：评论

学校读者我要写书评

暂无评论

Hybrid clustering strategies for effective oversampling and undersampling in multiclass classification

引用

SCIENTIFIC REPORTS 2025年第1期15卷 1-20页

作者： Salehi, Amirreza Khedmati, Majid Sharif Univ Technol Dept Ind Engn Tehran Iran Sharif Univ Technol Dept Ind Engn Azadi Ave Tehran *** Iran

multiclass imbalance is a challenging problem in real-world datasets, where certain classes may have a low number of samples because they correspond to rare occurrences. To address the challenge of multiclass imbalance, this paper introduces a novel hybrid cluster-based oversampling and undersampling (HCBOU) technique. By clustering and separating classes into majority and minority categories, this algorithm retains the most information during undersampling while generating efficient data in the minority class. The classification is carried out using one-vs-one and one-vs-all decomposition schemes. Extensive experimentation was carried out on 30 datasets to evaluate the proposed algorithm's performance. The results were subsequently compared with those of several state-of-the-art algorithms. Based on the results, the proposed algorithm outperforms the competing algorithms under different scenarios. Finally, The HCBOU algorithm demonstrated robust performance across varying class imbalance levels, highlighting its effectiveness in handling imbalanced datasets.

关键词： multiclass classification Imbalanced data Oversampling Undersampling Ensemble

来源：评论

学校读者我要写书评

暂无评论

multiclass classification BASED ON META PROBABILITY CODES

引用

INTERNATIONAL JOURNAL OF PATTERN RECOGNITION AND ARTIFICIAL INTELLIGENCE 2011年第8期25卷 1219-1241页

作者： Farajzadeh, Nacer Pan, Gang Wu, Zhaohui Yao, Min Zhejiang Univ Coll Comp Sci & Technol Hangzhou 310027 Zhejiang Peoples R China

This paper proposes a new approach to improve multiclass classification performance by employing Stacked Generalization structure and One-Against-One decomposition strategy. The proposed approach encodes the outputs of all pairwise classifiers by implicitly embedding two-class discriminative information in a probabilistic manner. The encoded outputs, called Meta Probability Codes (MPCs), are interpreted as the projections of the original features. It is observed that MPC, compared to the original features, has more appropriate features for clustering. Based on MPC, we introduce a cluster-based multiclass classification algorithm, called MPC-Clustering. The MPC-Clustering algorithm uses the proposed approach to project an original feature space to MPC, and then it employs a clustering scheme to cluster MPCs. Subsequently, it trains individual multiclass classifiers on the produced clusters to complete the procedure of multiclass classifier induction. The performance of the proposed algorithm is extensively evaluated on 20 datasets from the UCI machine learning database repository. The results imply that MPC-Clustering is quite efficient with an improvement of 2.4% overall classification rate compared to the state-of-the-art multiclass classifiers.

关键词： multiclass classification classifier stacked generalization decomposition one-against-one support vector machine multilayer perceptron clustering self organizing map

来源：评论

学校读者我要写书评

暂无评论

multiclass classification of distributed memory parallel computations

引用

PATTERN RECOGNITION LETTERS 2013年第3期34卷 322-329页

作者： Whalen, Sean Peisert, Sean Bishop, Matt Columbia Univ Dept Comp Sci New York NY 10027 USA Univ Calif Berkeley Lawrence Berkeley Natl Lab Berkeley CA 94720 USA Univ Calif Davis Dept Comp Sci Davis CA 95616 USA

High Performance Computing (HPC) is a field concerned with solving large-scale problems in science and engineering. However, the computational infrastructure of HPC systems can also be misused as demonstrated by the recent commoditization of cloud computing resources on the black market As a first step towards addressing this, we introduce a machine learning approach for classifying distributed parallel computations based on communication patterns between compute nodes. We first provide relevant background on message passing and computational equivalence classes called dwarfs and describe our exploratory data analysis using self organizing maps. We then present our classification results across 29 scientific codes using Bayesian networks and compare their performance against Random Forest classifiers. These models, trained with hundreds of gigabytes of communication logs collected at Lawrence Berkeley National Laboratory, perform well without any a priori information and address several shortcomings of previous approaches. (C) 2012 Elsevier B.V. All rights reserved.

关键词： multiclass classification Bayesian networks Random forests Self-organizing maps High performance computing Communication patterns

来源：评论

学校读者我要写书评

暂无评论

multiclass classification machines with the complexity of a single binary classifier

引用

SIGNAL PROCESSING 2013年第5期93卷 1013-1026页

作者： Honeine, Paul Noumir, Zineb Richard, Cedric Univ Technol Troyes LM2S Inst Charles Delaunay CNRSUMR 6279 Paris France Univ Nice Sophia Antipolis Observ Cote Azur Lab Lagrange CNRSUMR 7293 Sophia Antipolis France

In this paper, we study the multiclass classification problem. We derive a framework to solve this problem by providing algorithms with the complexity of a single binary classifier. The resulting multiclass machines can be decomposed into two categories. The first category corresponds to vector-output machines, where we develop several algorithms. In the second category, we show that the least-squares classifier can be easily cast into a multiclass one-versus-all scheme, without the need to train multiple binary classifiers. The proposed framework shows that, while keeping the classification accuracy essentially unchanged, the computational complexity is orders of magnitude lower than those previously reported in the literature. This makes our approach extremely powerful and conceptually simple. Moreover, we study the coding of the multiclass labels, and demonstrate that several celebrated approaches are equivalent. These arguments are illustrated with experimentations on well-known benchmarks. (C) 2012 Elsevier B.V. All rights reserved.

关键词： multiclass classification Machine learning SVM One-versus-all Least-squares classification

来源：评论

学校读者我要写书评

暂无评论

multiclass classification of whole-body scintigraphic images using a self-defined convolutional neural network with attention modules

引用

MEDICAL PHYSICS 2021年第10期48卷 5782-5793页

作者： Lin, Qiang Cao, Chuangui Li, Tongtong Cao, Yongchun Man, Zhengxing Wang, Haijun Northwest Minzu Univ Sch Math & Comp Sci Lanzhou Peoples R China Northwest Minzu Univ Key Lab Streaming Data Comp Technol & Applicat Lanzhou Peoples R China Minist Educ Key Lab Chinas Ethn Languages & Informat Technol Lanzhou Peoples R China Gansu Prov Hosp Dept Nucl Med Lanzhou Peoples R China

Purpose A self-defined convolutional neural network is developed to automatically classify whole-body scintigraphic images of concern (i.e., the normal, metastasis, arthritis, and thyroid carcinoma), automatically detecting diseases with whole-body bone scintigraphy. Methods A set of parameter transformation operations are first used to augment the original dataset of whole-body bone scintigraphic images. A hybrid attention mechanism including the spatial and channel attention module is then introduced to develop a deep classification network, Dscint, which consists of eight weight layers, one hybrid attention module, two normalization modules, two fully connected layers, and one softmax layer. Results Experimental evaluations conducted on a set of whole-body scintigraphic images show that the proposed deep classification network, Dscint, performs well for automated detection of diseases by classifying the images of concerns, achieving the accuracy, precision, recall, specificity, and F-1 score of 0.9801, 0.9795, 0.9791, 0.9933, and 0.9792, respectively, on the test data in the augmented dataset. A comparative analysis of Dscint and several classical deep classification networks (i.e., AlexNet, ResNet, VGGNet, DenseNet, and Inception-v4) reveals that our self-defined network, Dscint, performs best on classifying whole-body scintigraphic images on the same dataset. Conclusions The self-defined deep classification network, Dscint, can be utilized to automatically determine whether a whole-body scintigraphic image is either normal or contains diseases of concern. Specifically, better performance of Dscint is obtained on images with lesions that are present in relatively fixed locations like thyroid carcinoma than those with lesions occurring in nonfixed locations of bone tissue.

关键词： attention mechanism bone scintigraphy convolutional neural network medical image analysis multiclass classification

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：