Fahad Alharbi - Academia.edu (original) (raw)
Uploads
Papers by Fahad Alharbi
Computational Intelligence and Neuroscience, 2021
Noise in training data increases the tendency of many machine learning methods to overfit the tra... more Noise in training data increases the tendency of many machine learning methods to overfit the training data, which undermines the performance. Outliers occur in big data as a result of various factors, including human errors. In this work, we present a novel discriminator model for the identification of outliers in the training data. We propose a systematic approach for creating training datasets to train the discriminator based on a small number of genuine instances (trusted data). The noise discriminator is a convolutional neural network (CNN). We evaluate the discriminator’s performance using several benchmark datasets and with different noise ratios. We inserted random noise in each dataset and trained discriminators to clean them. Different discriminators were trained using different numbers of genuine instances with and without data augmentation. We compare the performance of the proposed noise-discriminator method with seven other methods proposed in the literature using seve...
Renewable Energy, 2011
Presented are the results of a comparative analysis to identify abundant, non-toxic binary materi... more Presented are the results of a comparative analysis to identify abundant, non-toxic binary materials with potential applicability for photovoltaics. Materials other than the conventional Si, CdTe, and Cu(In,Ga)Se 2 (CIGS) are examined. The screening is based on the materials' bulk properties and a set of environmental, physical, and chemical criteria. The screening process is detailed and the properties and applicability of the screened materials are discussed.
Optical and Quantum Electronics, 2008
An explicit finite difference method (FDM) to solve the nonparabolic effective mass approximation... more An explicit finite difference method (FDM) to solve the nonparabolic effective mass approximation of Schrodinger wave equation (SWE) for arbitrary quantum wells (QWs) is presented. The explicit nature of the presented method and its sparse matrices allow fast computation for energy states in QWs. The nonparabolicity effects are considered explicitly without iteration. This in turn results in faster and more stable calculations. The method is used to study the nonparabolicity effects in energy states and states overlapping in asymmetric AlGaAs/GaAs QWs.
Computational Intelligence and Neuroscience, 2021
Noise in training data increases the tendency of many machine learning methods to overfit the tra... more Noise in training data increases the tendency of many machine learning methods to overfit the training data, which undermines the performance. Outliers occur in big data as a result of various factors, including human errors. In this work, we present a novel discriminator model for the identification of outliers in the training data. We propose a systematic approach for creating training datasets to train the discriminator based on a small number of genuine instances (trusted data). The noise discriminator is a convolutional neural network (CNN). We evaluate the discriminator’s performance using several benchmark datasets and with different noise ratios. We inserted random noise in each dataset and trained discriminators to clean them. Different discriminators were trained using different numbers of genuine instances with and without data augmentation. We compare the performance of the proposed noise-discriminator method with seven other methods proposed in the literature using seve...
Renewable Energy, 2011
Presented are the results of a comparative analysis to identify abundant, non-toxic binary materi... more Presented are the results of a comparative analysis to identify abundant, non-toxic binary materials with potential applicability for photovoltaics. Materials other than the conventional Si, CdTe, and Cu(In,Ga)Se 2 (CIGS) are examined. The screening is based on the materials' bulk properties and a set of environmental, physical, and chemical criteria. The screening process is detailed and the properties and applicability of the screened materials are discussed.
Optical and Quantum Electronics, 2008
An explicit finite difference method (FDM) to solve the nonparabolic effective mass approximation... more An explicit finite difference method (FDM) to solve the nonparabolic effective mass approximation of Schrodinger wave equation (SWE) for arbitrary quantum wells (QWs) is presented. The explicit nature of the presented method and its sparse matrices allow fast computation for energy states in QWs. The nonparabolicity effects are considered explicitly without iteration. This in turn results in faster and more stable calculations. The method is used to study the nonparabolicity effects in energy states and states overlapping in asymmetric AlGaAs/GaAs QWs.