The rise of low-cost drones, or Unmanned Aerial Vehicles (UAVs), has transformed industries like agriculture, disaster response, and infrastructure inspection. However, this increase poses considerable issues for safety, security, and airspace management. To address these challenges, radiofrequency (RF)-based approaches are emerging as potential options for real-time UAV identification and classification since they are non-intrusive and compatible with existing systems. This research offers a unique Artificial Neural Network (ANN) model for UAV classification that includes efficient preprocessing and feature extraction approaches. The suggested architecture reduces feature dimensionality using Principal Component Analysis (PCA) and improves classification performance with statistical and frequency-domain characteristics like Root Mean Square (RMS) and Zero-Crossing Rate (ZCR). Experimental evaluations of the DroneRF dataset show that the proposed ANN is superior, with classification accuracies of 99.46%, 98.52%, and 83.40% for 2-class, 4-class, and 10-class tasks, respectively. This work is a big step toward robust, real-time drone detection and classification systems.

Ad hoc protocols like AODV, DSDV, and OLSR have been studied in healthcare, but their application to CGM networks with real-time integrated Machine Learning (ML) presents unique challenges due to distinct traffic patterns and node configurations. This paper explores how CGM, ML, and communication protocols jointly enable continuous monitoring and timely intervention, where transmission reliability directly affects ML performance in anomaly detection. Using NS3, we evaluate AODV, DSDV, and OLSR based on throughput, delay, and packet delivery ratio (PDR) under varying network sizes and data rates. Results show AODV offers minimal delay and packet loss, enhancing ML accuracy, while DSDV and OLSR perform better in dense networks. We also identify challenges related to scalability and data security, highlighting directions for future optimization of CGM-ML systems through hybrid and energyefficient protocols.

Heart disease continues to be the foremost cause of mortality across both developed and developing nations, posing a critical public health challenge. This life-threatening condition often necessitates urgent hospitalization and rapid diagnosis in emergency departments to avert fatal outcomes. Such outcomes can arise from a diverse range of complications associated with heart disease, including cardiogenic shock, pulmonary edema, myocardial rupture, the progression of congestive heart failure, thromboembolism, asystole, and ventricular fibrillation. Although numerous studies have focused on predicting heart disease, relatively few have addressed the complexity of multi-label classification for this issue. In this paper, we propose an innovative framework that integrates a K-Nearest Neighbors-based imputation technique with a Gradient Boosting algorithm to classify multiple heart disease complications. The experimental findings reveal that our approach achieves superior accuracy compared to existing methods. Additionally, we present a comprehensive comparative analysis of the proposed data pre-processing methodology by applying various machine learning algorithms. Moreover, we provide a detailed comparison between our multi-label classification strategy and previously established approaches. A diverse set of evaluation metrics, including accuracy, precision, recall, F1-score, and Hamming loss, is utilized to demonstrate the robustness and effectiveness of the proposed framework.

The goal of our research is to develop a model that achieves higher accuracy in predicting stock price movements, enabling investors to make informed decisions on buying or selling stocks to maximize their benefits. We propose a novel stock price prediction framework that integrates deep learning techniques and sentiment analysis. To achieve more convincing results, we incorporated sentiment data derived from newspaper articles. Using FinBERT (Financial-BERT), we analyzed the sentiments embedded in these articles and incorporated them into our LSTM model. Our model achieved the highest testing accuracy of 65.2 % which is higher than the research results we found. These results highlight the significant role of sentiment in influencing stock price movements and demonstrate the performance of our LSTM-based framework.

Class imbalance remains a significant challenge in medical machine learning, particularly in genomic classification tasks involving rare cancer subtypes. In microRNA (miRNA) expression analysis, the underrepresentation of minority classes can bias model training and compromise the reliable classification of clinically relevant cancer subtypes. This study addresses the class imbalance problem in The Cancer Genome Atlas (TCGA) miRNA expression data for three kidney cancer subtypes: TCGA-KIRC, TCGA-KIRP, and TCGA-KICH. The original dataset comprises 1,642,113 miRNA–patient records distributed unevenly across the three subtypes, with 970,596, 547,371, and 124,146 records, respectively. To mitigate this imbalance, we propose three Wasserstein generative adversarial network with gradient penalty (WGAN-GP)-based dataset construction procedures: Progressive Balancing, Direct Balancing, and Pretrained Balancing. Progressive Balancing increases the representation of minority classes to match the majority class, producing a balanced dataset of 2,911,788 records, with 970,596 records per class. Ten machine learning classifiers were evaluated using the pre-balanced-pool protocol, under which synthetic records were generated before cross-validation. Random Forest achieved the highest weighted F1-score of 0.8788 under this protocol. To assess generalization more rigorously, we additionally conducted five-fold patient-level stratified grouped cross-validation, excluding patient identifiers and file-path information from all learning stages and restricting WGAN-GP augmentation to the training portion of each fold. Under this leakage-controlled protocol, Random Forest achieved a weighted F1-score of 49.84% ± 3.11% at the 50% balancing target. This lower score provides a more conservative estimate of performance on unseen patients than the pre-balanced-pool result. Furthermore, a fractional-balancing ablation using a standardized dataset of 15,000 samples identified the 50% balancing target as the best-performing evaluated augmentation level, suggesting a potential trade-off between augmentation volume and synthetic-data quality. For secure deployment, we incorporated a lightweight physical unclonable function (PUF)-based hardware security layer on the Nexys A7-100T field-programmable gate array (FPGA) to support device authentication and on-demand key generation. The hardware module exhibited an approximate latency of 350 clock cycles for PUF response generation, error correction, and hashing; dataset encryption/decryption and end-to-end inference latency were not benchmarked. Overall, this study presents a WGAN-GP-based framework for addressing class imbalance in kidney cancer miRNA classification and highlights the importance of patient-level leakage control when evaluating model generalization. However, the observed performance gap between the pooled and grouped protocols cannot be attributed to leakage alone, because the protocols also differ in training budget. Further validation against external balancing baselines and independent patient cohorts is required to establish the clinical generalizability of the proposed approach.