检索结果-内蒙古大学图书馆

MKD-Net: A Novel Neuro Evolutionary Approach for Blockchain-Based Secure Medical Image classification Using Multi-Kernel DLM

引用

IEEE ACCESS 2025年 13卷 29900-29913页

作者： Vishwakarma, Virendra P. Yadav, Abhay Kumar USICT Guru Gobind Singh Indraprastha Univ New Delhi 110078 India

In the era of digital healthcare, secure and accurate classification of medical images is of utmost importance, particularly in situations of data imbalance. Owing to advancements in the domain of nature-inspired optimization and human brain simulations, this paper presents a novel approach to a neuro-evolutionary algorithm for medical image classification. In the proposed methodology, we develop a non-iterative learning-based neural network that can handle class imbalance with the help of class-specific regularizations, named multi-kernel deep neural networks (MKD-Net). Multiple kernels are employed in the proposed MKD-Net for different data representations in the classification of heterogeneous data. The MKD-Net parameters are optimized using the Enhanced-JAYA algorithm. ResNet-50, known for its incomparable feature drawing ability, has been used in MKD-Net as an input image feature extractor. Experimental evaluations conducted on benchmark medical image datasets demonstrate that MKD-Net achieves superior classification performance, particularly in imbalanced datasets. Key metrics such as classification accuracy, sensitivity, specificity, precision, and F1-score consistently improve over traditional models. Specifically, the proposed MKD-Net obtained accurate classification on ATT and ALL-IDB1 datasets;97.44% and 97.46% accuracy on ALL-IDB2 and CoVid datasets respectively. Moreover, the integration with blockchain provide secure feature storage and also ensures data integrity, traceability, and enhanced privacy. By combining advanced feature extraction, optimization, and security mechanisms, MKD-Net sets a new benchmark in medical image classification, showcasing its potential for scalable and secure healthcare applications.

关键词： Accuracy Medical diagnostic imaging Feature extraction Blockchains Image classification Face recognition Deep learning Optimization Leukemia classification algorithms Data Imbalance medical image classification multi-kernel deep learning model neuro-evolution ResNet-50

来源：评论

学校读者我要写书评

暂无评论

Online Neyman-Pearson classification With Hierarchically Represented Models

引用

IEEE JOURNAL OF SELECTED TOPICS IN SIGNAL PROCESSING 2025年第3期19卷 478-490页

作者： Can, Basarbatu Pelvan, Soner Ozgun Ozkan, Huseyin ABN Amro Bank NV NL-1082 PP Amsterdam Netherlands Sabanci Univ Fac Engn & Nat Sci Elect Engn TR-34956 Istanbul Turkiye

We consider the statistical anomaly detection problem with regard to false alarm rate (or false positive rate, FPR) controllability, nonlinear modeling and computational efficiency for real-time processing. A decision theoretical solution can be formulated as Neyman-Pearson (NP) hypothesis testing (binary classification: anomaly/nominal). In this framework, we propose an ensemble NP classifier (Tree OLNP) that is based on a binary partitioning tree. Tree OLNP generates an ensemble of sample space partitions. Each partition corresponds to an online piecewise linear (hence nonlinear) expert classifier as a union of online linear NP classifiers (union of OLNPs). While maintaining a precise control over the FPR, Tree OLNP generates its overall prediction as a performance driven and time varying weighted combination of the experts. This provides a dynamical nonlinear modeling power in the sense that simpler (more powerful) experts receive larger weights early (late) in the data stream, which manages the bias-variance trade-off and mitigates overfitting/underfitting issues. We mathematically prove that, for any stream, Tree OLNP asymptotically performs at least as well as of the best expert in terms of the NP performance with a regret diminishing in the order O(1/root t) (t: data size). Our algorithm is computationally highly efficient since it is online and its complexity scales linearly with respect to both the data size and tree depth, and scales twice-logarithmic with respect to the number of experts. We experimentally show that Tree OLNP strongly outperforms the state-of-the-art alternative techniques.

关键词： Costs classification algorithms Data models Controllability Complexity theory Computational modeling Tuning Training Signal processing algorithms Real-time systems Anomaly detection neyman pearson classification false alarm rate online decision tree partitioning

来源：评论

学校读者我要写书评

暂无评论

Deep Learning for Frailty classification Using IMU Sensor Data: Insights From FRAILPOL Database

引用

IEEE SENSORS JOURNAL 2025年第2期25卷 3974-3981页

作者： Amjad, Arslan Szczeesna, Agnieszka Blaszczyszyn, Monika Sacha, Jerzy Sacha, Magdalena Feusette, Piotr Wolanski, Wojciech Konieczny, Mariusz Borysiuk, Zbigniew Khan, Basheir Silesian Tech Univ Fac Automat Control Dept Comp Graph Vis & Digital Syst PL-44100 Gliwice Poland Opole Univ Technol Fac Phys Educ & Physiotherapy Dept Phys Educ & Sport PL-45758 Opole Poland Univ Opole Univ Hosp Inst Med Sci Dept Cardiol PL-45040 Opole Poland Univ Opole Inst Med Sci Fac Med Dept Family Med & Publ Hlth PL-45040 Opole Poland Univ Opole Univ Hosp Inst Med Sci Dept Rehabil PL-45040 Opole Poland Univ Malaya Inst Math Sci Fac Sci Kuala Lumpur 50603 Malaysia

Frailty is a significant health issue among the elders that leads to adverse outcomes, such as disability, increased healthcare utilization, and diminished overall well-being. This results in significant economic and personal costs, adding a considerable burden on the healthcare system. Early frailty detection contributes to establish a sustainable society by addressing health challenges and enhancing healthcare systems to be more responsive to aging populations. The FRAILPOL dataset was assessed, which contains data of 682 elderly participants. The dataset includes recordings from five inertial measurement unit (IMU) sensors aimed at classifying individuals into frail, prefrail, or robust (nonfrail) categories. The IMU data, comprising accelerometer and gyroscope readings from gait, are directly input into the InceptionTime, a deep learning (DL) algorithm for the classification task. The InceptionTime algorithm achieved an average accuracy of 81% on the test data. For the critical early frailty stage (prefrail), it reported 81% precision, 81% recall, and an F1-score of 80%. These findings can provide a valuable diagnostic tool for identifying frailty in its early stages, which can significantly contribute to the mitigation of frailty progression. In addition, the relatively simple observation of an elderly person's gait can be an important social factor toward recognizing frailty syndrome.

关键词： Sensors Sensor phenomena and characterization Accelerometers classification algorithms Medical services Accuracy Gyroscopes Deep learning Databases Image sensors Deep learning (DL) frailty gait healthcare inertial measurement unit (IMU) sensor

来源：评论

学校读者我要写书评

暂无评论

Artificial Flora Algorithm-Based Feature Selection With Support Vector Machine for Cardiovascular Disease classification

引用

IEEE ACCESS 2025年 13卷 7293-7309页

作者： Asha, M. M. Ramya, G. Vellore Inst Technol Sch Comp Sci Engn & Informat Syst Vellore 632014 Tamil Nadu India

Accurately categorizing medical information is crucial for determining effective cardiac treatment options, especially as the volume of data grows and feature selection becomes increasingly challenging. This work proposes a model to identify the presence of Cardiovascular Disease based on various patient features, aiming to enhance prediction accuracy through a powerful feature selection method. This approach utilizes the Cleveland dataset by combining the Artificial Flora Optimization algorithm with the Support Vector Machine. The proposed algorithm functions as a meticulous gardener, selectively identifying the most significant features for heart disease prediction through an objective function. The model demonstrates impressive performance, achieving an accuracy of 96.63%, specificity of 95.73%, sensitivity of 97.74%, precision of 94.89%, and an F1-score of 96.29%. The model promises high-accuracy heart disease predictions by optimizing feature selection, potentially transforming clinical practice, and advancing research. The novel combination of the proposed technique holds significant potential for improving medical categorization and patient outcomes.

关键词： Heart Feature extraction classification algorithms Prediction algorithms Genetic algorithms Accuracy Artificial intelligence Particle swarm optimization Artificial flora optimization fitness function feature selection ROC analysis support vector machine cardiovascular disease support vector machine cardiovascular disease

来源：评论

学校读者我要写书评

暂无评论

AlgaeClass_Net: Optimizing Few-Shot Marine Microalgae classification With Multi-Scale Feature Enhancement Network

引用

IEEE ACCESS 2025年 13卷 16223-16237页

作者： Liu, Dan Yuan, Guihong Tan, Huachao Jiang, Yanbo Bi, Hai Cheng, Yuan Dalian Ocean Univ Coll Informat Engn Dalian 116000 Peoples R China Hangzhou Yunxi Smart Vis Technol Co Ltd Dalian 116000 Peoples R China Ningbo Inst Dalian Univ Technol Ningbo 315000 Peoples R China Dalian Ocean Univ Key Lab Environm Controlled Aquaculture Minist Educ Dalian 116000 Peoples R China

As the eutrophication of the water body becomes more and more serious, the algae in the water body grow in large quantities and eventually form harmful algal blooms, causing great harm to the marine ecosystem. Therefore, how to quickly and accurately identify algae and make precautions becomes the key to solving this problem. Currently, more than tens of thousands of microalgae species are known around the world, but publicly available data are sparse, many of the species are characterized similarly to each other, and it is currently challenging to train an effective classification model with limited data. Existing few-shot learning classification algorithms that utilize meta-learning for metrics can be a good solution to this problem. In this paper, an AlgaeClass_Net algorithm that combines an improved multi-scale feature fusion with a feature enhancement module is proposed for the fine-grained features of microalgae. Furthermore, it utilizes a metric learning approach to classify few-shot microalgae by calculating the distances between the feature vectors of samples in the query set and the feature vectors of samples in the support set. The experimental results showed that the method achieves 78.55% and 91.20% classification accuracies under different tasks of 5-way 1-shot and 5-way 5-shot, respectively, with an improvement of 3.51% and 4.97% on the suboptimal model, respectively. It provides new research ideas for the identification of marine microalgae and the development and utilization of marine renewable energy.

关键词： Measurement classification algorithms Algae Microscopy Data models Metalearning Feature extraction Aquaculture Marine ecosystems Marine microalgae few-shot learning multi-scale feature fusion feature enhancement metric learning

来源：评论

学校读者我要写书评

暂无评论

Efficient Hierarchical Hash-Based Multi-Field Packet classification With Fast Update for Software Switches

引用

IEEE ACCESS 2025年 13卷 28962-28978页

作者： Chang, Yeim-Kuan Lin, Yu-Hung Chang, Jia-Chang Li, Wenjun Hsieh, Sun-Yuan Natl Cheng Kung Univ Dept Comp Sci & Informat Engn Tainan 70101 Taiwan Peng Cheng Lab Network Commun Res Ctr Shenzhen 518066 Peoples R China Natl Chi Nan Univ Dept Comp Sci & Informat Engn Puli 54561 Taiwan

Packet classification that attracts wide interests in router designs is one of the most important functionalities to provide services such as QoS, packet forwarding, firewall, and VPN. With the expanding network and the rise of software-defined networks, software routers/switches are needed to enable a wide variety of network applications in physical or virtual networks. Existing packet classification schemes used in software switches had become difficult to support a fast classification process for large multi-field rulesets. In addition, it is necessary to have fast updates that are frequently performed. In this paper, we propose a hash-based partitioning scheme for a multi-field packet classification problem. The proposed design is a 4-level data structure based on hashing. It selects the suitable fields to construct the data structures so that we can quickly narrow down the search space to find matched rules. The hash table used in the proposed design is more effective than the existing search data structures. Our experimental results show that the proposed scheme performs best in all four-performance metrics of memory consumption, construction time, classification time, and update time compared with the six most recently proposed schemes, CutTSS, TabTree, TupleMerge, CutSplit, PartitionSort, and PSTSS. The average number of bytes required per rule is only 27-99% of that needed in these schemes. The construction time of our scheme is 2.06 to 327.82 times faster than these schemes. Most importantly, the proposed scheme is 1.92-16.65 and 2.22-4.54 times faster than these six schemes in throughput and update time, respectively.

关键词： IP networks Control systems Protocols Software Decision trees classification algorithms Memory management Software defined networking Field programmable gate arrays Computer science Packet classification hash table tuple space search linear search OpenFlow

来源：评论

学校读者我要写书评

暂无评论

classification Based on the Support Vector Machine for Determining Operational Targets for Controlling Electricity Usage With Conventional Meters: A Case Study of Industrial and Business Tariff Customers From PT PLN (Persero) Indonesia

引用

IEEE ACCESS 2025年 13卷 12388-12398页

作者： Arisona, Galih Taruna, Alief Pascal Irwanto, Dwi Bestari, Arif Bijak Juniawan, Wildan Bandung Inst Technol Fac Math & Nat Sci Computat Sci Study Program Bandung 40132 Indonesia PT PLN Persero Div Syst & Informat Technol Head Off Jakarta Jakarta 12160 Indonesia Bandung Inst Technol Fac Math & Nat Sci Bandung 40132 Indonesia PT PLN Icon Plus Applicat Serv Div PT PLN Persero Jakarta 12710 Indonesia

Electricity theft remains a significant challenge for PT PLN (Persero), Indonesia's primary electricity provider, serving over 89 million customers as of 2023. The study focuses on industrial and business tariff customers, using a dataset from 2019 to 2023, which includes monthly consumption data from PLN's postpaid customers across thirty operational units with the highest Electricity Use Control (P2TL) levels, covering customers with a maximum power of 6,600 VA. This approach differs from previous studies that rely on open or smart meter data, as this study uses conventional meters for data collection. In the dataset used for this research, losses from confirmed electricity theft amounted to approximately IDR 19 billion. This research aims to improve the detection of electricity theft through a machine learning-based model utilizing the Support Vector Machine (SVM) classification technique. The goal is to enhance the P2TL mechanism by accurately identifying potential targets for field verification. Various SVM kernels were tested, including Radial Basis Function (RBF), Linear, Polynomial (Poly), and Sigmoid, alongside classifiers such as SVM, Logistic Regression, Decision Tree, and Na & iuml;ve Bayes. Results show that the SVM model, particularly with the RBF kernel, achieves optimal performance, with balanced precision and recall, especially with 30 months of historical data. This optimized model contributes to improving PLN's operational efficiency, offering more accurate identification of electricity theft cases, leading to substantial financial savings by reducing losses from unpaid consumption. The findings offer practical benefits for reducing electricity theft and improving PLN's monitoring system, especially in industrial and business sectors.

关键词： Electricity Support vector machines Data models Accuracy Meters Electric potential Machine learning Kernel Smart meters classification algorithms classification conventional meters electricity theft machine learning support vector machine

来源：评论

学校读者我要写书评

暂无评论

Node classification in Networks via Simplicial Interactions

引用

IEEE TRANSACTIONS ON NEURAL NETWORKS AND LEARNING SYSTEMS 2025年 PP卷 PP页

作者： Koo, Eunho Lim, Tongseok Chonnam Natl Univ Dept Big Data Convergence Gwangju 61186 South Korea Purdue Univ Mitch Daniels Sch Business W Lafayette IN 47907 USA

In the node classification task, it is natural to presume that densely connected nodes tend to exhibit similar attributes. Given this, it is crucial to first define what constitutes a dense connection and to develop a reliable mathematical tool for assessing node cohesiveness. In this article, we propose a probability-based objective function for semi-supervised node classification that takes advantage of higher order networks' capabilities. The proposed function reflects the philosophy aligned with the intuition behind classifying within higher order networks, as it is designed to reduce the likelihood of nodes interconnected through higher order networks bearing different labels. In addition, we propose the stochastic block tensor model (SBTM) as a graph generation model designed specifically to address a significant limitation of the traditional stochastic block model (SBM), which does not adequately represent the distribution of higher order structures in real networks. We evaluate the objective function using networks generated by the SBTM, which include both balanced and imbalanced scenarios. Furthermore, we present an approach that integrates the objective function with graph neural network (GNN)-based semi-supervised node classification methodologies, aiming for additional performance gains. Our results demonstrate that in challenging classification scenarios-characterized by a low probability of homo-connections, a high probability of hetero-connections, and limited prior node information-models based on the higher order network outperform pairwise interaction-based models. Furthermore, the experimental results suggest that integrating our proposed objective function with existing GNN-based node classification approaches enhances the classification performance by efficiently learning higher order structures distributed in the network.

关键词： Linear programming classification algorithms Stochastic processes Indexes Graph neural networks Training Tensors Probability distribution Probabilistic logic Performance gain Clique higher order networks hypergraph node classification node interaction probabilistic objective function semi-supervised simplex

来源：评论

学校读者我要写书评

暂无评论

RSTHFS: A Rough Set Theory-Based Hybrid Feature Selection Method for Phishing Website classification

引用

IEEE ACCESS 2025年 13卷 68820-68830页

作者： Setu, Jahanggir Hossain Halder, Nabarun Islam, Ashraful Amin, M. Ashraful Independent Univ Ctr Computat & Data Sci Dhaka 1229 Bangladesh

Phishing is a pervasive form of cybercrime where malicious websites deceive users into revealing sensitive information, e.g., passwords and credit card details. Despite advances in cybersecurity, accurately detecting phishing websites remains challenging due to the absence of universally accepted identification parameters. This study introduces a novel feature selection method, Rough Set Theory-based Hybrid Feature Selection (RSTHFS), to enhance phishing website detection using Machine Learning (ML) techniques. Our approach was evaluated using three diverse datasets containing 2,456, 10,000, and 88,647 instances. The RSTHFS method demonstrated a significant improvement by maintaining an average accuracy rate of 95.48% while reducing the number of features by 69.11% on average. Performance was further assessed using three advanced classifiers: Light Gradient-Boosting Machine (LightGBM), Random Forest (RF), and Categorical Boosting (CatBoost), with CatBoost emerging as the most efficient, achieving the highest accuracy. Additionally, RSTHFS reduced the runtime by 61.43%, highlighting its efficiency. These findings indicate that RSTHFS is not only effective in identifying phishing websites but also accelerates ML processes, providing a reliable and swift approach to feature selection. This work contributes to the field by presenting a robust methodology that enhances the accuracy and speed of phishing detection systems.

关键词： Phishing Feature extraction Accuracy Support vector machines classification algorithms Runtime classification tree analysis Rough sets Radio frequency Principal component analysis Cyber security feature selection hybrid feature machine learning phishing phishing websites rough set theory RSTHFS

来源：评论

学校读者我要写书评

暂无评论

Encrypted Traffic classification Through Deep Domain Adaptation Network With Smooth Characteristic Function

引用

IEEE TRANSACTIONS ON NETWORK AND SERVICE MANAGEMENT 2025年第1期22卷 331-343页

作者： Tong, Van Dao, Cuong Tran, Hai-Anh Tran, Duc Binh, Huynh Thi Thanh Hoang-Nam, Thang Tran, Truong X. Hanoi Univ Sci & Technol Sch Informat & Commun Technol Hanoi 10000 Vietnam Hanoi Univ Civil Engn Fac Informat Technol Hanoi 11600 Vietnam Penn State Univ Penn State Harrisburg Sch Sci Engn & Technol Middletown PA 16802 USA

Encrypted network traffic classification has become a critical task with the widespread adoption of protocols such as HTTPS and QUIC. Deep learning-based methods have proven to be effective in identifying traffic patterns, even within encrypted data streams. However, these methods face significant challenges when confronted with new applications that were not part of the original training set. To address this issue, knowledge transfer from existing models is often employed to accommodate novel applications. As the complexity of network traffic increases, particularly at higher protocol layers, the transferability of learned features diminishes due to domain discrepancies. Recent studies have explored Deep Adaptation Networks (DAN) as a solution, which extends deep convolutional neural networks to better adapt to target domains by mitigating these discrepancies. Despite its potential, the computational complexity of discrepancy metrics, such as Maximum Mean Discrepancy, limits DAN's scalability, especially when applied to large datasets. In this paper, we propose a novel DAN architecture that incorporates Smooth Characteristic Functions (SCFs), specifically SCF-unNorm (Unnormalized SCF) and SCF-pInverse (Pseudo-inverse SCF). These functions are designed to enhance feature transferability in task-specific layers, effectively addressing the limitations posed by domain discrepancies and computational complexity. The proposed mechanism provides a means to efficiently handle situations with limited labeled data or entirely unlabeled data for new applications. The aim is to limit the target error by incorporating a domain discrepancy between the source and target distributions along with the source error. Two statistics classes, SCF-unNorm and SCF-pInverse, are used to minimize this domain discrepancy in traffic classification. The experimental results demonstrate that our proposed mechanism outperforms existing benchmarks in terms of accuracy, enabling real-time traffic c

关键词： Cryptography Training Telecommunication traffic Payloads classification algorithms Virtual private networks Feature extraction Adaptation models Accuracy Home automation Traffic classification deep learning deep adaptation network multi-kernel maximum mean discrepancy and smooth characteristic function test

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：