Since most classifiers are biased toward the dominant class, class imbalance is a challenging problem in machine learning. The most popular approaches to solving this problem include oversampling minority examples and undersampling majority examples. Oversampling may increase the probability of overfitting, whereas undersampling eliminates examples that may be crucial to the learning process. We present a linear time resampling method based on random data partitioning and a majority voting rule to address both concerns, where an imbalanced dataset is partitioned into a number of small subdatasets, each of which must be class balanced. After that, a specific classifier is trained for each subdataset, and the final classification result is established by applying the majority voting rule to the results of all of the trained models. We compared the performance of the proposed method to some of the most well-known oversampling and undersampling methods, employing a range of classifiers, on 33 benchmark machine learning class-imbalanced datasets. The classification results produced by the classifiers employed on the generated data by the proposed method were comparable to most of the resampling methods tested, with the exception of SMOTEFUNA, which is an oversampling method that increases the probability of overfitting. The proposed method produced results that were comparable to the Easy Ensemble (EE) undersampling method. As a result, for solving the challenge of machine learning from class-imbalanced datasets, we advocate using either EE or our method.

Biometric technology has received a lot of attention in recent years. One of the most prevalent biometric traits is the finger-knuckle print (FKP). Because the dorsal region of the finger is not exposed to surfaces, FKP would be a dependable and trustworthy biometric. We provide an FKP framework that uses the VGG-19 deep learning model to extract deep features from FKP images in this paper. The deep features are collected from the VGG-19 model’s fully connected layer 6 (F6) and fully connected layer 7 (F7). After applying multiple preprocessing steps, such as combining features from different layers and performing dimensionality reduction using principal component analysis (PCA), the extracted deep features are put to the test. The proposed system’s performance is assessed using experiments on the Delhi Finger Knuckle Dataset employing a variety of common classifiers. The best identification result was obtained when the Artificial neural network (ANN) classifier was applied to the principal components of the averaged feature vector of F6 and F7 deep features, with 95% of the data variance preserved. The findings also demonstrate the feasibility of employing these deep features in an FKP recognition system.

Augmented reality (AR) has shown potential for aiding users in their assembly tasks. It provides a magic-lens view of the physical object. Thus, assembly tasks become less demanding. An assembly task using AR provides a pre-defined sequence of actions with a minimum amount of information to do the task. Thus, there is a need to exploit the advances in AR technology to provide assemblers with a more sophisticated learning experience. Using AR to turn the physical object into a smart object that communicates and interacts with a user can result in an assembler achieving a more knowledge and a greater awareness of the technology and provide insights and a deeper understanding of the software components inside the embedded computing instead of only learning the assembly hardware components. This study used a Pedagogical Virtual Machine (PVM) to evaluate the learning effectiveness with AR. We presented an experimental work based on assembling and exploring a modularized mobile robot task called Buzz-Boards. A total of 36 students participated in the study; 18 students participated in both experimental and control groups. Findings demonstrated that the PVM with AR approach was more effective in learning achievement, learning activity enjoyment, and usefulness than the paper-based approach.

Artificial intelligence (AI), particularly deep learning, is reshaping breast cancer diagnostics in the radiology and pathology fields. This review synthesizes recent advances in mammography, digital breast tomosynthesis (DBT), ultrasound, MRI, and whole-slide imaging, with an emphasis on convolutional neural networks (CNNs), Vision Transformers (ViTs), and generative adversarial networks (GANs). When embedded within established screening and diagnostic workflows, AI systems can enhance lesion detection and triage, as well as reduce interpretive variability. However, performance and generalizability depend on dataset quality, population and vendor heterogeneity, acquisition protocols, and calibrated probability outputs; diminished performance on external datasets and miscalibration remain recurrent risks that require explicit mitigation during development and deployment of these models. Beyond detection and classification, segmentation and risk prediction models increasingly integrate imaging with clinicopathological and, where available, genomic variables to enable individualized risk stratification and follow-up planning. Data generation strategies, including GAN-based augmentation, can partially address data scarcity and class imbalance but require rigorous quality control and bias monitoring. Persistent barriers to clinical adoption include uneven external validation, domain shifts across institutions, variability in reporting standards, limited interpretability, and ethical, privacy, and regulatory constraints. Overall, AI should augment, rather than replace, the role of clinicians. Priorities for responsible integration include multi-site prospective evaluations, transparent and standardized reporting, bias mitigation, robust calibration, and lifecycle monitoring to ensure sustained safety and equity.