Scopus Research — Zeyad Taha Yaseen

In recent years, financial institutions have increasingly relied on artificial intelligence (AI) to evaluate credit ratings and economic risk. In the meantime, Many artificial intelligence-oriented models operate as black boxes, making decision-making difficult to understand. Lack of transparency raises doubts regarding regulatory compliance, justice, and responsibility. Explainable artificial intelligence (Xai) solves these challenges when one understands model predictions. The popular Xai method used to help grasp complex artificial intelligence models to ensure that their results are more dependable and understandable is Shapley's (SAPL) additive explanations. This paper investigates the application of the SHAP in evaluating financial risks and credit scores. Real-world financial data sets to train machine learning models, including gradient reinforcement techniques, neural networks, and decision-making trees. After that, the artificial intelligence decision-making process is investigated using the SHAP concerning several factors like income, credit history, and job status. This work evaluates SHAP's performance in improving interpretability and justice. Initial findings suggest that the SHAP provides unambiguous justifications for resource value and significantly increases the transparency of the model. Improved interpretability helps financial companies identify likely biases, enhancing ratings and regulatory compliance. According to the studies, AI financial models increase accountability and confidence. Including artificial intelligence-oriented credit scores into their operations would help most open financial companies improve decision-making processes, follow policies, and increase customer confidence. © 2025 IEEE.

The integration of optical fibre communication with multiple input multiple output-non-orthogonal multiple access (MIMO-NOMA) waveforms in cognitive radio (CR) systems is examined in this study. The proposed system leverages the advantages of optical fibre, including high bandwidth and immunity to electromagnetic interference to facilitate the transmission and reception of MIMO-NOMA signals in a CR environment. Moreover, MIMO-NOMA signal was detected and analysed by the hybrid-discrete cosine transform-Welch (H-DCT-W) method. Based on the modes results, a detection probability greater than 0.96%, a false alarm probability equal to 0.06, and a global system error probability equal to 0.09% were obtained with a signal-to-noise ratio (SNR) less than 0 dB, while maintaining a simple level of complexity. The results obtained in this paper indicate the potential of the optical fibre-based MIMO-NOMA system based on H-DCT-W technology in CR networks. Therefore, its suitability for practical CR applications is demonstrated by the improvements obtained in false alarms, detection probability, and error rates at low levels of SNR. This study contributes to the development of efficient and reliable wireless communication systems by linking cooperation and synergy concerning MIMO-NOMA, optical fibres, as well as the proposed detection technique (H-DCT-W). © 2024 The Author(s).

In the rapidly evolving landscape of smart cities, the integration of advanced technologies is crucial for ensuring safety, optimizing traffic flow, and enhancing the urban living experience. Vehicle-to-vehicle (V2V) communication and visible light communication (VLC) have emerged as promising solutions to address these challenges. This paper explores the integration of V2V communication and VLC at smart pedestrian crosswalks to enhance pedestrian safety and traffic management in smart cities. It explores the impact of neighbouring vehicles on V2V-VLC performance and proposes novel methodologies to assess traffic density effects. Results indicate a significant chance of encountering nearby cars during rush hours, emphasizing the importance of these integrated systems for safety and mobility in urban environments. The outcomes show that the chance of running into extra cars in nearby lanes is independent of the particular lane and increases to 80% through rush hours, but falls to a lower amount than 20% through off-peak and initial morning hours. © 2024 The Author(s).

Using no-core optical fiber (NCF) and conventional single-mode fiber (SMF), this research needs to increase the sensitivity of temperature and relative humidity measurements for radiology device applications. Operating on the Mach-tender Interferometer (MZI) principle, the sensors make use of surface plasmon resonance. Traditional sensing technologies sometimes confront difficulties in reaching high sensitivity and accuracy, especially in demanding environments like radiology facilities. In order to increase the sensitivity of four relative humidity sensors constructed from SMF, TNCF, and SMF, a 2 cm NCF segment was incorporated. To enhance sensitivity, the sensors have been coated with a variety of substances, such as gold nanoparticles (AuNPs), silica nanoparticles (PVN/PVP), and a layer-by-layer technique. The sensor coated with AuNPS exhibited the greatest sensitivity with respect to temperature, attaining 0.486 db/°C. The fiber optic sensors that have been developed exhibit great potential for utilization in biomedical and environmental domains. Experimental tests are carried out to assess the performance of the proposed sensing system, including sensitivity, accuracy, and reaction time. © 2024 Author(s).