There are many features that have been taken into consideration for wind power forecasting. Since properly ranking these relevant features, often redundant, can be quite difficult, highly accurate short term wind power forcasting remains a big challenge. Another noted problem that adversely impacts the accuracy of wind forcasting stems from the weakness of the prevailing prediction models based on the feedforward neural network (FNN) in handling wind power time series. This paper thus attempts to address the aforementioned problems in short-term wind power forecasting with a novel approach that combines the infinite feature selection (Inf-FS) with the recurrent neural networks (RNN). In particular, all the possible features related to wind forecast are first clustered into multiple feature sets, after which the identified feature sets are mapped onto the paths of a graph built for Inf-FS. Traversing such a graph helps effectively determine/rank the significance of the features according to their stability and classification accuracy measured in the feature space. The proposed wind prediction model then feeds the ranked features into a deep learning prediction system enabled by RNN, whose neurons have self-feedback loops to help gather the past decisions, and thus be more effective than FNN for wind power prediction. The proposed wind power prediction approach is demonstrated through the experimental evaluations using a dataset from the National Renewable Energy Laboratory (NREL). The result shows that the accuracy of short-term wind power forecast is increased by 11%, 29%, 33%, and 19% in spring, summer, autumn and winter, respectively, over that achieved using the traditional approaches.
As we advance in time, with the evolution of the computer industry and the escalating intricacy of diverse software, the demand for source code defects is also increasing. Traditional deep learning-based software defect detection methods perform well on synthetic defect datasets, but their performance on real software defect datasets is unsatisfactory. At the same time, pre-trained models derived from extensive data training are extensively employed in various NLP tasks and have achieved excellent results. Based on this background, this study introduces a software defect detection system leveraging CodeBERT and Bi-LSTM. This method first preprocesses and standardizes the C source code to minimize the impact of redundant information and reduce noise; Secondly, coding sequences are segmented and encoded using the CodeBERT pre-trained model, capturing the semantic features within the program's code and converting them into vectors containing code feature information and output; Then, the output of the received CodeBERT is processed through the Bi-LSTM network to acquire the structural composition inherent to the code and the semantic information of the positive and negative terms; Finally, the vectors containing source code features are classified using a fully connected network to determine whether the code segment has defects. To verify the impact of this approach, we used the cross-language benchmark test set CodeXGLUE proposed by Microsoft Research Institute for evaluating code tasks for validation. The results showed that this method had higher accuracy in detecting real software defect samples than other methods, indicating that the proposed method can effectively improve software defect detection capabilities.
1 January 2021 in Computer Modeling in Engineering & Sciences
The COVID-19 has resulted in catastrophic situation and the deaths of millions of people all over the world. In this paper, the predictions of epidemiological propagation models, such as SIR and SEIR, are introduced to analyze the earlier COVID-19 propagation. The deep learning methods combined with transfer learning are familiar with classification-detection approaches based on chest X-ray and CT images are presented in detail. Besides, deep learning approaches have also been applied to lung ultrasound (LUS), which has been shown to be more sensitive than chest X-ray and CT images in detecting COVID-19. In the absence of a vaccine, the machine learning-related approaches are applied to analyze vaccine candidates in the realm of biology and medicine. The telehealth system played a major role in combating the pandemic from all aspects and reducing contact with patients during this period. Natural language processing-related methods are utilized to analyze tweets related to the COVID-19 epidemic on social media, and further analyze public sentiment and subject modeling, so as to arrange corresponding measures to appease public sentiment. In particular, this survey is to summarize and analyze the contributions made in various fields during the COVID-19 pandemic by considering both the contribution of deep learning in chest X-ray and CT images, as well as the application of the latest LUS during the COVID-19 pandemic. Telehealth and the importance of public sentiment analysis during a pandemic were also described in detail.
1 January 2025 in American Journal of Translational Research
Accurate classification of skin cancer is critical for early detection and timely treatment, significantly improving patient survival rates. While quantum neural networks combined with transfer learning show promise in medical image analysis, quantum noise remains a major challenge, compromising the stability and reliability of these systems. This study aims to address this limitation by developing a robust quantum-based framework for skin cancer classification. We propose a Quantum Dual-Branch Neural Network (QDBNN) that employs two independently trained network branches without shared weights. Dual-modal features are fused at the fully connected layer, and a Variational Quantum Classifier (VQC) is utilized for final classification. The model is evaluated on two datasets: the multiclass HAM10000 and the binary Malignant vs. Benign dataset. QDBNN achieved state-of-the-art accuracies of 93.6% on HAM10000 and 93.5% on the Malignant vs. Benign dataset, outperforming classical and quantum transfer learning baselines. The dual-branch architecture and weighted feature fusion demonstrated enhanced robustness against quantum noise while improving generalization. QDBNN effectively mitigates quantum noise interference and leverages quantum-classical hybrid advantages for skin cancer classification. Its success highlights the potential of quantum-inspired architectures in medical imaging, offering a pathway toward clinically deployable tools for early diagnosis. Future work will focus on hardware optimization and scalability to larger datasets.
26 June 2023 in Journal of Biomolecular Structure and Dynamics
Notwithstanding the extensive research efforts directed towards devising a dependable approach for the diagnosis of coronavirus disease 2019 (COVID-19), the inherent complexity and capriciousness of the virus continue to pose a formidable challenge to the precise identification of affected individuals. In light of this predicament, it is essential to devise a model for COVID-19 prediction utilizing chest computed tomography (CT) scans. To this end, we present a hybrid quantum-classical convolutional neural network (HQCNN) model, which is founded on stochastic quantum circuits that can discern COVID-19 patients from chest CT images. Two publicly available chest CT image datasets were employed to evaluate the performance of our model. The experimental outcomes evinced diagnostic accuracies of 99.39% and 97.91%, along with precisions of 99.19% and 98.52%, respectively. These findings are indicative of the fact that the proposed model surpasses recently published works in terms of performance, thus providing a superior ability to precisely predict COVID-19 positive instances.Communicated by Ramaswamy H. Sarma
The inherent randomness, intermittence and volatility of wind power generation compromise the quality of the wind power system, resulting in uncertainty in the system's optimal scheduling. As a result, it's critical to improve power quality and assure real-time power grid scheduling and grid-connected wind farm operation. Inferred statistics are utilized in this research to infer general features based on the selected information, confirming that there are differences between two forecasting categories: Forecast Category 1 (0–11 h ahead) and Forecast Category 2 (12–23 h ahead). In z-tests, the null hypothesis provides the corresponding quantitative findings. To verify the final performance of the prediction findings, five benchmark methodologies are used: Persistence model, LMNN (Multilayer Perceptron with LM learning methods), NARX (Nonlinear autoregressive exogenous neural network model), LMRNN (RNNs with LM training methods) and LSTM (Long short-term memory neural network). Experiments using a real dataset show that the LSTM network has the highest forecasting accuracy when compared to other benchmark approaches including persistence model, LMNN, NARX network, and LMRNN, and the 23-steps forecasting accuracy has improved by 19.61%.