相关参考文献
注意：仅列出部分参考文献，下载原文获取全部文献信息。
Article
Computer Science, Information Systems
Xinmin Tao et al.
Summary: The imbalanced dataset classification problem is a major challenge in the field of machine learning. This paper proposes a novel SVDD boundary-based weighted oversampling approach (SVDDWSMOTE) to handle imbalanced and overlapped data. The approach generates the class boundary using the support vector data description (SVDD) model and eliminates potential overlapped or noisy samples. It also addresses the within-balance issues by assigning weights based on densities and distances to the SVDD class boundary. The proposed approach achieves significant improvements in classification performance compared to state-of-the-art methods.
INFORMATION SCIENCES
(2022)
Article
Computer Science, Information Systems
Yuting Sun et al.
Summary: In this paper, a hypergraph regularized semi-supervised support vector machine (HGSVM) algorithm is proposed to exploit the multivariate manifold structure. To accelerate the training process of HGSVM, a fast algorithm based boundary sample selection algorithm, termed fast-HGSVM, is developed. Moreover, two SMOTE-variant techniques and the one-vs-rest strategy are introduced in fast-HGSVM, and two multi-category semi-supervised algorithms called fast-ASHGSVM and fast-KSHGSVM are proposed. The effectiveness of the proposed algorithms is validated through experiments on two moons and UCI datasets.
INFORMATION SCIENCES
(2022)
Article
Automation & Control Systems
Xiaona Song et al.
Summary: This article investigates the issue of adaptive neural network fixed-time tracking control for a class of strict-feedback nonlinear systems with prescribed performance demands. By incorporating an improved fractional-order command filtered backstepping control technique and an event-driven-based fixed-time adaptive controller, the signals of the closed-loop system are practically fixed-time bounded and the tracking error is regulated.
JOURNAL OF THE FRANKLIN INSTITUTE-ENGINEERING AND APPLIED MATHEMATICS
(2022)
Article
Chemistry, Multidisciplinary
Xiaorou Zheng et al.
Summary: Hyperspectral remote sensing image classification is an important method widely used in various fields. However, the imbalance in the dataset during classification may result in biased and inaccurate results. This paper proposes a preprocessing solution called NearPseudo based on semi-supervised learning, which adds pseudo-labels to minority classes and improves accuracy through a consistency check mechanism. Experimental results demonstrate the effectiveness of NearPseudo in improving classification accuracy.
APPLIED SCIENCES-BASEL
(2022)
Article
Computer Science, Information Systems
Der -Chiang Li et al.
Summary: Learning from class-imbalanced data is challenging, and using synthetic minority oversampling technique (SMOTE) has shown improvement in classifying minority class. However, SMOTE and its extensions may generate noise instances and suffer from performance degradation. To address these issues, a new algorithm named RIOT is proposed, which can locate and identify hard-to-learn minority instances based on region radius and class ratio, and generate synthetic instances until balance is approximately achieved.
INFORMATION SCIENCES
(2022)
Article
Computer Science, Information Systems
Nan Wang et al.
Summary: In this paper, a search-based cost-sensitive hypergraph learning method is proposed for anomaly detection. The method effectively addresses the issues of class imbalance and unclear correlation commonly encountered in anomaly detection tasks. Experimental results on industry datasets and other benchmark datasets demonstrate the effectiveness and superiority of the proposed method.
INFORMATION SCIENCES
(2022)
Article
Mathematics, Applied
Vladimir Djordjevic et al.
Summary: This paper addresses the control problem of hydraulic servo actuators (HSA) with unknown dynamics, utilizing an adaptive dynamic programming approach via output feedback. An online learning data-driven controller based on measured input and output data is used, eliminating the need for knowledge of the HSA dynamics or exosystem dynamics. The theoretical convergence of the ADP based control algorithm is demonstrated, with simulation results confirming the feasibility and effectiveness of the proposed approach in solving the optimal control problem of HSA.
DISCRETE AND CONTINUOUS DYNAMICAL SYSTEMS-SERIES S
(2022)
Article
Engineering, Mechanical
Kun Yu et al.
Summary: A three-stage semi-supervised learning method is proposed in this study for intelligent bearing fault diagnosis under limited labeled data, using data augmentation and metric learning to enhance classifier performance. Experimental results demonstrate that the proposed method outperforms existing diagnostic methods in bearing fault diagnosis under limited labeled samples.
MECHANICAL SYSTEMS AND SIGNAL PROCESSING
(2021)
Article
Computer Science, Information Systems
Wei Feng et al.
Summary: An adaptive semi-supervised rotation forest (SSRoF) algorithm is proposed for the classification of hyperspectral images with limited training data, which increases the number of training instances by mining high-quality unlabeled samples and utilizes SMOTE to overcome class imbalance. The effectiveness of the proposed method is demonstrated on three real hyperspectral remote sensing datasets through comparisons with ensemble methods and semi-supervised methods.
INFORMATION SCIENCES
(2021)
Article
Computer Science, Information Systems
Ming Zheng et al.
Summary: A novel three-stage undersampling framework, UFFDFR, is proposed to improve classification performance on imbalanced data by removing noise and unrepresentative samples. Experiments show that UFFDFR outperformed classic and state-of-the-art clustering-based undersampling methods in terms of F-measure, G-mean, and AUC for five classification algorithms.
INFORMATION SCIENCES
(2021)
Article
Computer Science, Artificial Intelligence
Remi Viola et al.
Summary: This paper addresses the challenging problem of imbalanced learning from a nearest-neighbor classification perspective, introducing an algorithm that rescales distances to modify decision boundaries. The proposed method aims to reduce False Negative rate while controlling False Positives, showing effectiveness compared to popular Nearest-Neighbor algorithms and state-of-the-art sampling methods in extensive experimental studies.
INTERNATIONAL JOURNAL ON ARTIFICIAL INTELLIGENCE TOOLS
(2021)
Article
Computer Science, Information Systems
Giuseppina Andresini et al.
Summary: In this study, a new intrusion detection method is introduced which leverages a deep metric learning methodology combining autoencoders and Triplet networks. Two separate autoencoders are trained on historical normal network flows and attacks, and a Triplet network is trained to learn the embedding of the feature vector representation of network flows. This methodology achieves better predictive accuracy in detecting new signs of malicious activities in network traffic compared to competitive intrusion detection architectures on benchmark datasets.
INFORMATION SCIENCES
(2021)
Article
Computer Science, Artificial Intelligence
Cunjun Wang et al.
Summary: This paper proposes a novel deep metric learning model to address imbalanced data distribution and quickly recognize new faults in fault diagnosis. Experimental results demonstrate that the proposed model can automatically extract distinctive features, balance data representation, and perform open-set classification effectively.
KNOWLEDGE-BASED SYSTEMS
(2021)
Article
Computer Science, Artificial Intelligence
Jianjin Deng et al.
Summary: The study introduces a simple graph-based semi-supervised learning method that can effectively address various degrees of class imbalance and enhance the discriminative power of graph-based learning through estimating class proportions. Experimental results demonstrate promising performance of the approach in handling classification problems.
PATTERN RECOGNITION
(2021)
Article
Computer Science, Information Systems
Junnan Li et al.
Summary: Class imbalance is a significant factor leading to performance deterioration in classifiers. Techniques such as SMOTE and its extension, NaNSMOTE, have been successful in addressing this issue and have been proven effective on real data sets.
INFORMATION SCIENCES
(2021)
Article
Computer Science, Information Systems
Xiaocui Li et al.
WORLD WIDE WEB-INTERNET AND WEB INFORMATION SYSTEMS
(2020)
Article
Computer Science, Artificial Intelligence
Jesper E. Van Engelen et al.
Article
Computer Science, Artificial Intelligence
Man Leung Wong et al.
EXPERT SYSTEMS WITH APPLICATIONS
(2020)
Proceedings Paper
Computer Science, Artificial Intelligence
Wanyu Lin et al.
2020 IEEE/CVF CONFERENCE ON COMPUTER VISION AND PATTERN RECOGNITION (CVPR)
(2020)
Review
Multidisciplinary Sciences
Mahmut Kaya et al.
Article
Computer Science, Artificial Intelligence
Mateusz Buda et al.
Article
Computer Science, Artificial Intelligence
Anil K. Jain
PATTERN RECOGNITION LETTERS
(2010)
