Article
Computer Science, Artificial Intelligence
Bhagat Singh Raghuwanshi
Summary: The conventional extreme learning machine (ELM) fails to handle the class imbalance problem effectively because it treats all samples as equally important. To address this issue, modified versions of ELM like weighted ELM (WELM) and overall distribution WELM (ODW-ELM) have been developed. In this study, a class-specific ELM based on overall distribution (OD-CSELM) and its kernelized version (OD-CSKELM) are proposed to handle binary class imbalance problem more effectively. OD-CSELM and OD-CSKELM have lower computational complexity compared to WELM and kernelized WELM, respectively. Experimental results on benchmark datasets demonstrate the superior generalization performance of the proposed methods for class imbalance learning.
Article
Computer Science, Artificial Intelligence
Bhagat Singh Raghuwanshi, Sanyam Shukla
Summary: This paper introduces a new variant of extreme learning machine, MCVCSELM, for effectively addressing binary class imbalance problems by utilizing minimum class variance and class-specific regularization. Experimental results demonstrate that the proposed algorithm outperforms several state-of-the-art methods for imbalanced learning.
EXPERT SYSTEMS WITH APPLICATIONS
(2021)
Article
Computer Science, Artificial Intelligence
Qiude Li, Qingyu Xiong, Shengfen Ji, Yang Yu, Chao Wu, Min Gao
Summary: MIS-ELM is an Incremental Semi-supervised Extreme Learning Machine designed for Mixed data stream classification, which tackles the challenges of mixed data streams and limited labeled samples with innovative encoding and incremental learning methods. Experimental results demonstrate the superiority of the proposed method in real data streams.
EXPERT SYSTEMS WITH APPLICATIONS
(2021)
Article
Computer Science, Artificial Intelligence
Jiongming Qin, Cong Wang, Qinhong Zou, Yubin Sun, Bin Chen
Summary: Active learning combined with Extreme Learning Machine can reduce the cost of labeling instances and improve learning efficiency. The proposed AI-WSELM framework in this paper effectively handles multiclass imbalanced data and stream-based data, showing satisfactory performance compared to existing models.
KNOWLEDGE-BASED SYSTEMS
(2021)
Article
Computer Science, Artificial Intelligence
Adrian Rubio-Solis, Uriel Martinez-Hernandez, Luciano Nava-Balanzar, Luis G. Garcia-Valdovinos, Noe A. Rodriguez-Olivares, Juan P. Orozco-Muniz, Tomas Salgado-Jimenez
Summary: The paper proposes the use of Online Interval Type-2 Fuzzy Extreme Learning Machine (OIT2-FELM) for robust following behavior along a predefined 3D path using a Remotely Operated Underwater Vehicle (ROV). The OIT2-FELM offers better treatment of uncertainty and noisy signals underwater while improving ROV performance, with approximately 90.5% accuracy in testing data for classification predictions.
APPLIED SOFT COMPUTING
(2022)
Article
Computer Science, Information Systems
Purvi Prajapati, Amit Thakkar
Summary: Extreme Multi-Label Classification (XMLC) is a specific case of Multi-Label Classification that deals with a large number of labels. The main objective is to learn a classifier that can extract the relevant subset of labels from an extremely large label space. In an extreme environment, the performance of the classifier is affected by the large number of features, labels, and instances. The proposed approach, KTXMLC, utilizes a tree-based classifier and input representation technique to maintain correlations between features and labels. It outperforms existing tree-based classifiers in terms of ranking-based measures on multiple datasets.
JOURNAL OF KING SAUD UNIVERSITY-COMPUTER AND INFORMATION SCIENCES
(2022)
Article
Computer Science, Information Systems
Jiandong Zhou, Fengshi Jing, Xuejin Liu, Xiang Li, Qingpeng Zhang
Summary: In this study, a field-aware attentive neural factorization machine (FAFM) model is proposed for large-scale data-driven company investment valuation. The FAFM model outperforms existing baselines in prediction accuracy and model interpretability, providing a useful tool for investors to make better investment decisions.
INFORMATION SCIENCES
(2022)
Article
Computer Science, Artificial Intelligence
Audi Albtoush, Manuel Fernandez-Delgado, Eva Cernadas, Senen Barro
Summary: Quick Extreme Learning Machine (QELM) is a method that can handle large classification datasets by avoiding tuning and replacing training patterns in the activation matrix with a reduced set of prototypes, thus avoiding the storage and computation of large matrices. It can be executed on general purpose computers within reasonable times and achieves performances similar to extreme learning machine (ELM).
NEURAL COMPUTING & APPLICATIONS
(2022)
Article
Computer Science, Artificial Intelligence
Roshani Choudhary, Sanyam Shukla
Summary: This paper proposes an ensemble method that decomposes a complex imbalanced problem into simpler sub-problems, solves them using cost-sensitive classifiers, and combines the results using voting methods.
EXPERT SYSTEMS WITH APPLICATIONS
(2021)
Article
Computer Science, Information Systems
Xiaopeng Zhang, Liangxi Qin
Summary: In this paper, a new extreme learning machine algorithm called OWA-ELM is proposed to improve the classification performance of imbalanced data. Experimental results show that the OWA-ELM algorithm achieves better results in dealing with imbalanced data classification.
Article
Computer Science, Artificial Intelligence
Zijia Zhang, Yaoming Cai, Wenyin Gong
Summary: This paper presents a novel semi-supervised learning framework, Graph Convolutional Extreme Learning Machines (GCELM), for handling graph data in non-Euclidean domains. The proposed methods achieve significantly better results than previous methods on 36 benchmark datasets, thanks to the use of random graph convolution and a voting ensemble strategy.
EXPERT SYSTEMS WITH APPLICATIONS
(2023)
Article
Computer Science, Artificial Intelligence
Lin Sun, Tianxiang Wang, Weiping Ding, Jiucheng Xu
Summary: This study develops a novel Partial Multilabel Learning (PML) model that addresses some issues in traditional PML models by introducing fuzzy neighborhood-based ball clustering and kernel extreme learning machine (KELM). The model preprocesses the data with ball k-means clustering, designs a new ball clustering model, develops the particle-ball fusion strategy, studies fuzzy membership functions and label enhancement, and constructs a nonsmooth convex objective function. Experimental results on 14 datasets confirm the effectiveness of the proposed algorithm.
IEEE TRANSACTIONS ON FUZZY SYSTEMS
(2023)
Article
Computer Science, Artificial Intelligence
Bhagat Singh Raghuwanshi, Sanyam Shukla
Summary: This study proposes two novel methods, MVKWELM and MVCSKELM, for handling imbalanced classification problems more effectively. These methods enhance the algorithm's generalization capability through minimum variance embedding, with MVCSKELM combining the advantages of minimum variance embedding and class-specific regularization parameters.
PATTERN RECOGNITION
(2021)
Article
Computer Science, Artificial Intelligence
Jiahua Luo, Chi-Man Wong, Chi-Man Vong
Summary: Sparse Bayesian extreme learning machine (SBELM) faces challenges in multi-class classification, leading to the proposal of multinomial Bayesian extreme learning machine (MBELM) with multinomial distribution and integration of automatic relevance determination (ARD) and L1 penalty mechanisms. Experimental results demonstrate significant improvements in accuracy and model size for the new model.
Review
Computer Science, Artificial Intelligence
Xiulin Zheng, Peipei Li, Xindong Wu
Summary: This paper provides a comprehensive review of Extreme Learning Machine (ELM) theoretical research and its variants in data stream classification, introducing the basic principles, different variants, optimization strategies, practical applications, and experimental results. The open issues and prospects of ELM models for stream classification are also discussed.