Results 31 to 40 of about 93,524 (286)

Hybrid oversampling technique for imbalanced pattern recognition: Enhancing performance with Borderline Synthetic Minority oversampling and Generative Adversarial Networks

Machine Learning with Applications
Class imbalance problems (CIP) are one of the potential challenges in developing unbiased Machine Learning models for predictions. CIP occurs when data samples are not equally distributed between two or multiple classes.
Md Manjurul Ahsan, Shivakumar Raman, Yingtao Liu, Zahed Siddique   +3 more
doaj   +1 more source

Oversampling for Imbalanced Learning Based on K-Means and SMOTE

, 2017
Learning from class-imbalanced data continues to be a common and challenging problem in supervised learning as standard classification algorithms are designed to handle balanced class distributions. While different strategies exist to tackle this problem,
Bacao, Fernando, Douzas, Georgios, Last, Felix   +2 more
core   +1 more source

Fractional biorthogonal partners in channel equalization and signal interpolation [PDF]

, 2002
The concept of biorthogonal partners has been introduced recently by the authors. The work presented here is an extension of some of these results to the case where the upsampling and downsampling ratios are not integers but rational numbers, hence, the ...
Vaidyanathan, P. P., Vrcelj, Bojan
core   +3 more sources

Adaptive equalization in oversampled subbands [PDF]

Conference Record of Thirty-Second Asilomar Conference on Signals, Systems and Computers (Cat. No.98CH36284), 1998
The potential presence of fractional delays, nonminimum phase parts, and a colouring of the channel output can require adaptive equalisers to adapt very long filters, which can have slow convergence for LMS-type adaptive algorithms. The authors present a novel oversampled subband approach to adaptive equalisation, which can both significantly reduce ...
Weiß, S., Dooley, S.R., Stewart, R.W., Nandi, A.K.   +3 more
openaire   +5 more sources

A systematic study of the class imbalance problem in convolutional neural networks

, 2018
In this study, we systematically investigate the impact of class imbalance on classification performance of convolutional neural networks (CNNs) and compare frequently used methods to address the issue.
Buda, Mateusz, Maki, Atsuto, Mazurowski, Maciej A.   +2 more
core   +1 more source

Computational Multiscale Methods for Linear Poroelasticity with High Contrast [PDF]

, 2018
In this work, we employ the Constraint Energy Minimizing Generalized Multiscale Finite Element Method (CEM-GMsFEM) to solve the problem of linear heterogeneous poroelasticity with coefficients of high contrast.
Altmann, Robert, Chung, Eric T., Fu, Shubin, Maier, Roland, Peterseim, Daniel, Pun, Sai-Mang   +5 more
core   +2 more sources

Multiscale-Spectral GFEM and optimal oversampling [PDF]

Computer Methods in Applied Mechanics and Engineering, 2020
In this work we address the Multiscale Spectral Generalized Finite Element Method (MS-GFEM) developed in [I. Babuška and R. Lipton, Multiscale Modeling and Simulation 9 (2011), pp. 373--406]. We outline the numerical implementation of this method and present simulations that demonstrate contrast independent exponential convergence of MS-GFEM solutions.
Ivo Babuska, Robert Lipton 0001, Paul Sinz, Michael Stuebner   +3 more
openaire   +3 more sources

Handling Class Imbalanced Data in Sarcasm Detection with Ensemble Oversampling Techniques

Applied Artificial Intelligence
The rise of social media has amplified online sharing, necessitating businesses to comprehend public sentiment. Traditional sentiment analysis struggles with sarcasm detection and class imbalance.
Ya-Han Hu, Ting-Hsuan Liu, Chih-Fong Tsai, Yu-Jung Lin   +3 more
doaj   +1 more source

Digital Self-Interference Cancellation Based on Blind Source Separation and Spectral Efficiency Analysis for the Full-Duplex Communication Systems

IEEE Access, 2018
In this paper, the fundamental problem for the full-duplex communication systems, i.e., self-interference cancellation (SIC), is investigated, and a novel digital-domain SIC method based on blind source separation is proposed. This method achieves SIC by
Hua Yang, Hang Zhang, Jiang Zhang, Liu Yang   +3 more
doaj   +1 more source

The Real-World-Weight Cross-Entropy Loss Function: Modeling the Costs of Mislabeling

IEEE Access, 2020
In this paper, we propose a new metric to measure goodness-of-fit for classifiers: the Real World Cost function. This metric factors in information about a real world problem, such as financial impact, that other measures like accuracy or F1 do not. This
Yaoshiang Ho, Samuel Wookey
doaj   +1 more source