Search

Dealing with imbalanced data is a common challenge in machine learning, where one class has significantly fewer examples than another. Successfully addressing this challenge requires careful consideration of the data, algorithm, and evaluation metrics to ensure that the model accurately predicts the minority class. In this study, we present a hybrid approach called Siamese‐HYNAA, which combines a Siamese network and a population‐based optimizer hypercube natural aggregation algorithm (HYNAA) to generate candidate solutions for augmenting the minority class. We collected 10 imbalanced datasets ranging from 1.81 to 8.78 imbalanced ratios and built solution pairs based on correctly predicted candidate solutions using support vector machine (SVM). We then fed these solutions to the Siamese network, which employs a one‐shot learning approach to improve predictions with fewer candidate solutions. However, we found that SVM predicted only a small number of minority class samples accurately, prompting us to optimize the number of candidate solution pairs using HYNAA to generate more synthetic samples for the Siamese network. We evaluated our proposed strategy against basic SMOTE and our previous work, SMOTE‐PSOEV, using various performance measures, including ROC‐AUC learning curves, sensitivity, specificity, accuracy, Characteristic stability index, balanced accuracy, F1‐score, informedness, markedness, and execution time. Our results indicate that Siamese‐HYNAA generates promising results for imbalanced data. [ABSTRACT FROM AUTHOR]

As the "Internet of Medical Things (IoMT)" grows, healthcare systems can collect and process data. It is also challenging to study public health prevention requirements. Virus transmission can be prevented by wearing a mask. The World Health Organization (WHO) recommends wearing a facemask to protect against the COVID‐19 pandemic—the levels of a pandemic rise across almost all regions of the world. By following the WHO rules, we support the development of face mask‐detecting technologies and determine whether or not people are using masks in public locations. The proposed paradigm in this paper will work in three stages. Firstly, we use an Image data generator to import the images. In addition to using a Haar cascade (HC) classifier for detecting faces, residual learning (ResNet152V2) trains a model that detects whether someone is wearing a face mask. Detection and classification are carried out in real‐time with high precision. Compared with other recently proposed methods, the model achieved 99.65% accuracy during training and 99.63% during validation. [ABSTRACT FROM AUTHOR]