Hamoud Aljamaan

Android malware categories exhibit distinct behaviors, yet existing studies typically evaluate one ensemble strategy or feature representation, providing limited evidence of cross-category stability. This study compares a standalone decision tree, six homogeneous ensembles, and three heterogeneous ensembles across Adware, Banking malware, SMS malware, Riskware, and Combined Malware detection using static, dynamic, and hybrid features from CICMalDroid2020. With hybrid features, Stacking achieved mean Matthews correlation coefficient (MCC) values of 0.971–0.996 across the five tasks. Stacking and Soft Voting had the lowest descriptive average ranks, at 2.7 and 2.9, while several boosting ensembles achieved competitive MCC values. Adware and SMS malware were easier to detect than Banking malware and Riskware. Category-specific detectors achieved higher mean MCC in all 40 ratio-matched comparisons, with greater target-category training exposure. Hybrid features had the lowest descriptive average rank across the 50 model–task settings (1.28), followed by static (1.92) and dynamic (2.80) features. Their observed benefit over static features was strongest for Banking and Combined Malware. For hybrid features, Stacking’s median fitting time was about 5.5 times that of voting, while batch inference medians were close. The results describe observed ensemble performance and consistency, category-specific training outcomes, feature-representation differences, and computational trade-offs. These findings are limited to CICMalDroid2020 under a fixed 2:1 benign-to-malware ratio, and generalizability to other datasets and deployment conditions remains to be established.

Federated Learning (FL) enables decentralized model training across multiple clients while preserving data privacy; however, it faces significant challenges related to statistical heterogeneity and fairness across participating clients. Prior work has shown that dominant clients can disproportionately influence global model updates, leading to performance imbalance, and that manual intervention, such as removing high-performing clients, can partially improve fairness at the cost of automation and scalability. In this study, we propose an adaptive proximal regulation framework that automatically adjusts the proximal regularization term in the federated proximal algorithm to promote fair client participation during training. Unlike fixed regularization approaches, our method dynamically modifies the proximal coefficient based on client performance at each communication round. We evaluate two adaptive strategies under both independent and non-independent data distributions. Experimental results reveal that directly penalizing dominant clients does not consistently improve fairness and may degrade performance in heterogeneous environments. Motivated by this finding, we introduce a reverse adaptive strategy that applies stronger regularization to low-performing clients to reduce excessive local bias. This approach improves early-stage fairness and maintains competitive global accuracy. Our results highlight that effective fairness control in FL requires careful treatment of straggler clients rather than simple suppression of dominant ones. The proposed framework provides a step toward fully automated, fairness-aware federated optimization in real-world heterogeneous settings.

Investments are the foundation of financial markets, directing resources toward activities that generate future returns. Stock trading plays a central role in wealth creation, yet forecasting stock prices remains difficult due to the nonlinear and volatile nature of financial data. This study presents an ensemble deep learning framework for short-term stock price forecasting in the Saudi market. It addresses the limited use of advanced ensemble methods in emerging economies. The framework uses multivariate data from ten companies listed on the Tadawul All Share Index and compares two ensemble strategies: averaging and stacking with standalone models (Gated Recurrent Unit, Long Short-Term Memory, and Bidirectional Recurrent Neural Network). A total of 34 engineered technical indicators and a rolling-window setup were applied for five-day-ahead predictions. The results show that ensemble models consistently outperform individual architectures across all evaluation metrics. Stacking ensembles, especially those combining Gated Recurrent Unit and Long Short-Term Memory, achieved the highest accuracy. Averaging ensembles provided stable and efficient alternatives. Overall, the findings confirm the robustness and adaptability of ensemble methods for financial forecasting and offer practical insights for investors and policymakers seeking reliable, data-driven prediction tools.

Smart Contract (SC) vulnerabilities are programming errors or design flaws that can lead to financial loss or functional failure, making accurate detection essential. Although Machine Learning (ML) is widely applied to SC vulnerability detection, existing datasets are often small, imbalanced, inconsistently labeled, or nonstandardized, and frequently rely on limited feature representations that do not account for different contract lifecycle stages, restricting generalization and degrading benchmark reliability. This study introduces DIVE, a multi-label dataset that addresses these structural and feature-level limitations. DIVE includes 22,330 real-world SCs deployed between 2016 and 2024, and spanning major Solidity compiler versions, annotated for eight vulnerability types aligned with the Decentralized Application Security Project (DASP) Top 10 taxonomy. It provides 221 pre-deployment and 176 post-deployment features and employs a standardized multi-tool labeling pipeline based on Power-based voting and post-hoc filtering, which corrected 14.3% false positives in DoS and 24.9% in Time Manipulation. Unlike prior datasets, DIVE offers two lifecycle-specific feature sets and an open-source framework enabling reproducible benchmarking and periodic reconstruction aligned with evolving vulnerability patterns.

Smart Contracts (SCs), self-executing programs on blockchain platforms, are transforming industries such as banking, healthcare, and supply chains through automated, trustless transactions. However, their inherent vulnerabilities have led to severe financial and operational losses, with large-scale exploits causing substantial economic damage. Machine Learning (ML) has emerged as a promising approach for SC vulnerability detection, yet its effectiveness, adaptability, and generalizability remain insufficiently explored. This article comprehensively classifies current Ethereum SC vulnerabilities and attacks. It also surveys 108 ML-based detection methods, covering both traditional models and a structured taxonomy of advanced approaches such as GNN-based, LLM-based, contrastive learning, ensemble, hybrid, meta-learning, and transfer learning techniques. The strengths, limitations, and practical challenges of these methods are systematically analyzed, with particular attention to factors such as detection stages, classification problems, dataset characteristics, feature engineering, performance evaluation, generalizability, detection capability, model aging, and ethical and privacy implications. Additionally, existing datasets on SC vulnerabilities are reviewed and consolidated. By integrating these insights, this work provides actionable guidelines and a foundation for building secure, resilient, and trustworthy SC ecosystems.

Halal tourism is a rapidly growing industry that caters to the needs of Muslim travelers worldwide. This field’s research is also becoming more intriguing since it offers products such as halal restaurants, sharia hotels, banking, medicines, and finance. With the increasing popularity of halal tourism, it is essential to understand the sentiments of tourists regarding halal tourism products and services. Sentiment analysis is a useful tool to extract the polarity of reviews and opinions about products or services. In this paper, we present a survey of existing literature on sentiment analysis in halal tourism and the use of machine learning techniques or extracting sentiment polarity from reviews. We discuss various machine learning algorithms, including Random Forest, Naive Bayes, Support Vector Machine, Decision Tree, Convolutional Neural Network, BERT, and attention-based models, used for sentiment analysis in halal tourism. Our survey also covers various aspects of halal tourism, including food, cosmetics, hotels, and finance sectors. This paper highlights the importance of sentiment analysis in halal tourism and provides insight for future research.

Code smells are code structures that indicate a potential issue in code design or implementation. These issues could affect the processes of code testing and maintenance, and overall software quality. Therefore, it is important to detect code smells in the early stages of software development to enhance system quality. Most studies have focused on detecting code smells of a single programming language. This article explores TL for cross-language code smell detection, where Java is the source, and both C# and Python are the target datasets, focusing on Large Class, Long Method, and Long Parameter List code smells. We conducted a comparison study across two transfer learning approaches—instance-based (Importance Weighting Classifier, Nearest Neighbors Weighting, and Transfer AdaBoost) and parameter-based (Transfer Tree, Transfer Forest)—with various base models. The results showed that the instance-based approach outperformed the parameter-based approach, particularly with Transfer AdaBoost using ensemble learning base models. The Transfer AdaBoost approach with Gradient Boosting and Extra Trees achieved consistent and robust results across both C# and Python, with an 83% winning rate, as indicated by the Wilcoxon signed-rank test. These findings underscore the effectiveness of transfer learning for cross-language code smell detection, supporting its generalizability across different programming languages.