۱.
Online reviews are crucial in influencing consumer decisions and business practices. However, some individuals exploit this system by posting fake reviews, known as spam opinions, to manipulate perceptions. Spam detection systems face significant challenges in robustness due to their primary focus on identifying spam attacks without accounting for adversaries that target the detection mechanisms. This oversight enables spammers to exploit vulnerabilities in traditional algorithms with complex deceptive strategies, ultimately undermining their effectiveness. This paper proposes a novel multi-layer graph-based method that represents reviews, reviewers, and products as interconnected nodes. This approach captures the complex relationships among them and addresses adversarial attempts to manipulate the detection process. Our approach utilizes three key nodes—opinion, reviewer, and product—to assess the honesty, trust, and reliability of reviews, reviewers, and products in the context of potential deception. Furthermore, we develop a simulation tool capable of generating diverse attack scenarios, including those targeting the detection system itself, enabling a comprehensive evaluation of robustness. We compared the performance of our method with other graph-based techniques through simulations and case studies, demonstrating that our method is a competitive solution among existing alternatives.
۲.
Objective: This review explores the trustworthiness of multimodal artificial intelligence (AI) systems, specifically focusing on vision-language tasks. It addresses critical challenges related to fairness, transparency, and ethical implications in these systems, providing a comparative analysis of key tasks such as Visual Question Answering (VQA), image captioning, and visual dialogue. Background: Multimodal models, particularly vision-language models, enhance artificial intelligence (AI) capabilities by integrating visual and textual data, mimicking human learning processes. Despite significant advancements, the trustworthiness of these models remains a crucial concern, particularly as AI systems increasingly confront issues regarding fairness, transparency, and ethics. Methods: This review examines research conducted from 2017 to 2024, focusing on forenamed core vision-language tasks. It employs a comparative approach to analyze these tasks through the lens of trustworthiness, underlining fairness, explainability, and ethics. This study synthesizes findings from recent literature to identify trends, challenges, and state-of-the-art solutions. Results: Several key findings were highlighted. Transparency: The explainability of vision language tasks is important for user trust. Techniques, such as attention maps and gradient-based methods, have successfully addressed this issue. Fairness: Bias mitigation in VQA and visual dialogue systems is essential for ensuring unbiased outcomes across diverse demographic groups. Ethical Implications: Addressing biases in multilingual models and ensuring ethical data handling is critical for the responsible deployment of vision-language systems. Conclusion: This study underscores the importance of integrating fairness, transparency, and ethical considerations in developing vision-language models within a unified framework.
۳.
Software-Defined Network (SDN) introduces centralized network control via the OpenFlow protocol, enhancing network management, traffic routing, and security policy enforcement. However, SDN's centralized nature also introduces vulnerabilities, particularly to cyberattacks targeting the controller and communication channels. This study presents a resilience assessment methodology for SDN under cyberattack conditions, leveraging Markov process theory to model system states and transitions. Three SDN architectures were evaluated under various attack scenarios, revealing that traditional configurations lack sufficient resilience against synchronous attacks and controller breaches. To address these vulnerabilities, we propose an enhanced SDN protection framework integrating controller redundancy, automatic reconfiguration mechanisms, and anomaly detection using Long Short-Term Memory (LSTM) networks. The methodology was validated through simulations in the EVE-NG environment, demonstrating improved SDN stability under cyber threats. These findings provide a foundation for designing more resilient SDN infrastructures, ensuring network continuity and security against evolving cyber threats.
۴.
IoT is a dynamic network of interconnected things that communicate and exchange data, where security is a significant issue. Previous studies have mainly focused on attack classifications and open issues rather than presenting a comprehensive overview on the existing threats and vulnerabilities. This knowledge helps analyzing the network in the early stages even before any attack takes place. In this paper, the researchers have proposed different security aspects and a novel Bayesian Security Aspects Dependency Graph for IoT (BSAGIoT) to illustrate their relations. The proposed BSAGIoT is a generic model applicable to any IoT network and contains aspects from five categories named data, access control, standard, network, and loss. This proposed Bayesian Security Aspect Graph (BSAG) presents an overview of the security aspects in any given IoT network. The purpose of BSAGIoT is to assist security experts in analyzing how a successful compromise and/or a failed breach could impact the overall security and privacy of the respective IoT network. In addition, root cause identification of security challenges, how they affect one another, their impact on IoT networks via topological sorting, and risk assessment could be achieved. Hence, to demonstrate the feasibility of the proposed method, experimental results with various scenarios has been presented, in which the security aspects have been quantified based on the network configurations. The results indicate the impact of the aspects on each other and how they could be utilized to mitigate and/or eliminate the security and privacy deficiencies in IoT networks.
۵.
Generating dynamic videos from static images and accurately modeling object motion within scenes are fundamental challenges in computer vision, with broad applications in video enhancement, photo animation, and visual scene understanding. This paper proposes a novel hybrid framework that combines convolutional neural networks (CNNs), recurrent neural networks (RNNs) with long short-term memory (LSTM) units, and generative adversarial networks (GANs) to synthesize temporally consistent and spatially realistic video sequences from still images. The architecture incorporates splicing techniques, the Lucas-Kanade motion estimation algorithm, and a loop feedback mechanism to address key limitations of existing approaches, including motion instability, temporal noise, and degraded video quality over time. CNNs extract spatial features, LSTMs model temporal dynamics, and GANs enhance visual realism through adversarial training. Experimental results on the KTH dataset, comprising 600 videos of fundamental human actions, demonstrate that the proposed method achieves substantial improvements over baseline models, reaching a peak PSNR of 35.8 and SSIM of 0.96—representing a 20% performance gain. The model successfully generates high-quality, 10-second videos at a resolution of 720×1280 pixels with significantly reduced noise, confirming the effectiveness of the integrated splicing and feedback strategy for stable and coherent video generation.
۶.
In today’s data-driven hospitality sector, customer interactions increasingly occur through digital platforms, generating extensive behavioral and transactional information. This study analyse the prediction of Customer Lifetime Value (CLV) using machine learning models—Linear Regression, Random Forest, and LightGBM—trained on features derived from hotel website interactions and booking records. After comprehensive data preprocessing, the models were evaluated using MAE, RMSE, and R² metrics. LightGBM achieved the highest predictive performance (R² = 0.504), followed by Random Forest (R² = 0.497), while Linear Regression underperformed (R² = 0.386), highlighting the advantages of non-linear models in modeling intricate customer patterns. Residual analyses confirmed LightGBM's stability and low bias across diverse customer profiles. Apart from prediction, the study applies Recency-Frequency-Monetary (RFM) analysis to segment customers into distinct value-based groups. These segments form the basis for tailored marketing strategies, allowing hotels to allocate resources more efficiently, enhance customer retention, and develop targeted campaigns aligned with customer potential. By integrating web-derived behavioral data with advanced modeling and segmentation, this research offers hotel managers practical tools for strategic planning in customer relationship management.
