Hamada Nayel
Current Organizations / Institutes
The organizations/institutes from the scholar's latest article
Research Subjects
Hamada Nayel has a total of 161 co-authors. Here are the top co-authors by number of shared publications:
Nsrin Ashraf
Department of Computer Science Faculty of Computers and Artificial Intelligence Benha University Egypt
9 shared publicationsMariam Labib
Computer Engineering Elsewedy University of Technology Cairo Egypt
5 shared publicationsMohamed Taha
Department of Computer Science, Faculty of Computers and Artificial intelligence, Benha University, Benha, Egypt
4 shared publicationsMohammed Aldawsari
Department of Computer Engineering and Information, College of Engineering Wadi Al-Dawasir, Prince Sattam Bin Abdulaziz University, Al-Kharj 16273, Saudi Arabia
4 shared publications- Ahmed A. El-Sawy
Department of Computer Science, Faculty of Computers and Artificial intelligence, Benha University, Benha, Egypt
3 shared publications - Eman Shehab
Department of Computer Science, Faculty of Computers and Artificial Intelligence, University of Sadat City, Sadat, Egypt
3 shared publications
Latest Publications
- Open Access
- 6 May 2026
- Published by MDPI AG in Sustainability
- Vol. 18 (9) , 4617
Explainable AI-Driven Health Scoring Framework for Smart City Sustainability
The rapid evolution of smart cities demands a transition from reactive healthcare systems to proactive, data-driven health management paradigms that support long-term urban sustainability. Predicting population health status based on lifestyle-related behavioral and physiological factors is critical for enabling early intervention, personalized healthcare, and efficient resource allocation directly contributing to the United Nations Sustainable Development Goals (SDG 3: Good Health and Well-being; SDG 11: Sustainable Cities and Communities). This study proposes an IoT-enabled Explainable Artificial Intelligence (XAI) framework for predictive health scoring as part of sustainable population health management, integrating real-time data acquisition, cloud-based analytics, and interpretable machine learning. To address the limitations of conventional ensemble models particularly the black-box nature and hyperparameter sensitivity of Extreme Gradient Boosting (XGBoost) a Bayesian optimization strategy is employed to automatically fine-tune model parameters, thereby enhancing predictive accuracy and generalization performance. Furthermore, Shapley Additive Explanations (SHAP) are incorporated to provide transparent, interpretable insights into model predictions by quantifying the contribution of individual lifestyle features. Using a publicly available Kaggle dataset (“Health and Lifestyle Data for Regression”), experimental evaluation demonstrates that the proposed Bayesian-Optimized XGBoost model achieves superior performance (Test R2 = 0.878, RMSE = 4.983), outperforming ten benchmark models, including standard XGBoost, which exhibits signs of overfitting (Test R2 = 0.832). The results further reveal that Body Mass Index (BMI) and diet quality are the most influential factors affecting health scores, providing actionable insights for urban health policymakers. The proposed framework highlights the synergy between IoT, optimization techniques, and explainable AI to develop transparent, reliable, and scalable predictive health systems. This work provides a practical foundation for next-generation smart healthcare applications and decision-support systems, advancing the vision of sustainable, data-driven, and human-centric smart cities.
- 6 May 2026
- Published by MDPI AG in Sustainability
- Vol. 18 (9) , 4617
Open AccessExplainable AI-Driven Health Scoring Framework for Smart City Sustainability
The rapid evolution of smart cities demands a transition from reactive healthcare systems to proactive, data-driven health management paradigms that support long-term urban sustainability. Predicting population health status based on lifestyle-related behavioral and physiological factors is critical for enabling early intervention, personalized healthcare, and efficient resource allocation directly contributing to the United Nations Sustainable Development Goals (SDG 3: Good Health and Well-being; SDG 11: Sustainable Cities and Communities). This study proposes an IoT-enabled Explainable Artificial Intelligence (XAI) framework for predictive health scoring as part of sustainable population health management, integrating real-time data acquisition, cloud-based analytics, and interpretable machine learning. To address the limitations of conventional ensemble models particularly the black-box nature and hyperparameter sensitivity of Extreme Gradient Boosting (XGBoost) a Bayesian optimization strategy is employed to automatically fine-tune model parameters, thereby enhancing predictive accuracy and generalization performance. Furthermore, Shapley Additive Explanations (SHAP) are incorporated to provide transparent, interpretable insights into model predictions by quantifying the contribution of individual lifestyle features. Using a publicly available Kaggle dataset (“Health and Lifestyle Data for Regression”), experimental evaluation demonstrates that the proposed Bayesian-Optimized XGBoost model achieves superior performance (Test R2 = 0.878, RMSE = 4.983), outperforming ten benchmark models, including standard XGBoost, which exhibits signs of overfitting (Test R2 = 0.832). The results further reveal that Body Mass Index (BMI) and diet quality are the most influential factors affecting health scores, providing actionable insights for urban health policymakers. The proposed framework highlights the synergy between IoT, optimization techniques, and explainable AI to develop transparent, reliable, and scalable predictive health systems. This work provides a practical foundation for next-generation smart healthcare applications and decision-support systems, advancing the vision of sustainable, data-driven, and human-centric smart cities.
- 21 April 2026
- Published by Institute of Electrical and Electronics Engineers (IEEE)
- p. 1-6
RAG-Enhanced Healthcare Integration LLM Framework
With the revolution of Artificial Intelligence (AI), Large Language Models (LLMs) have demonstrated great capabilities in healthcare transformation through clinical question answering, medical decision support, and patient communication. However, the deployment of LLMs in real-world healthcare environments faces many challenges such as hallucination, lack of transparency, data privacy, and insufficient accountability. The paper proposes to use Retrieval-Augmented Generation (RAG) and developed a theoretical RAG-Enhanced healthcare LLM framework to address these limitations through component integration architecture. The proposed framework integrates contextual retrieval, confidence aware scoring, provenance tagging, and clinical guardrails with mandatory human-in-the-loop oversight. The framework reduces the risks of hallucination and other limitations of LLMs. The empirical assessment of the framework’s impact on diagnostic accuracy, clinical decision-making, and patient outcomes is the future work of this study.
- 21 April 2026
- Published by Institute of Electrical and Electronics Engineers (IEEE)
- p. 1-6
RAG-Enhanced Healthcare Integration LLM Framework
With the revolution of Artificial Intelligence (AI), Large Language Models (LLMs) have demonstrated great capabilities in healthcare transformation through clinical question answering, medical decision support, and patient communication. However, the deployment of LLMs in real-world healthcare environments faces many challenges such as hallucination, lack of transparency, data privacy, and insufficient accountability. The paper proposes to use Retrieval-Augmented Generation (RAG) and developed a theoretical RAG-Enhanced healthcare LLM framework to address these limitations through component integration architecture. The proposed framework integrates contextual retrieval, confidence aware scoring, provenance tagging, and clinical guardrails with mandatory human-in-the-loop oversight. The framework reduces the risks of hallucination and other limitations of LLMs. The empirical assessment of the framework’s impact on diagnostic accuracy, clinical decision-making, and patient outcomes is the future work of this study.
- Open Access
- 20 March 2026
- Published by Elsevier in Procedia Computer Science
- Vol. 275, 618-627
A rule-based model with Fuzzy matching for disease named entity recognition in Arabic biomedical texts
- 20 March 2026
- Published by Elsevier in Procedia Computer Science
- Vol. 275, 618-627
Open AccessA rule-based model with Fuzzy matching for disease named entity recognition in Arabic biomedical texts
- Open Access
Transformer-Based Model for Multilingual Hope Speech Detection
This paper describes a system that has been submitted to the "PolyHope-M" at RANLP2025. In this work various transformers have been implemented and evaluated for hope speech detection for English and Germany. RoBERTa has been implemented for English, while the multilingual model XLM-RoBERTa has been implemented for both English and German languages. The proposed system using RoBERTa reported a weighted f1-score of 0.818 and an accuracy of 81.8% for English. On the other hand, XLM-RoBERTa achieved a weighted f1-score of 0.786 and an accuracy of 78.5%. These results reflects the importance of improvement of pre-trained large language models and how these models enhancing the performance of different natural language processing tasks.
Open AccessTransformer-Based Model for Multilingual Hope Speech Detection
This paper describes a system that has been submitted to the "PolyHope-M" at RANLP2025. In this work various transformers have been implemented and evaluated for hope speech detection for English and Germany. RoBERTa has been implemented for English, while the multilingual model XLM-RoBERTa has been implemented for both English and German languages. The proposed system using RoBERTa reported a weighted f1-score of 0.818 and an accuracy of 81.8% for English. On the other hand, XLM-RoBERTa achieved a weighted f1-score of 0.786 and an accuracy of 78.5%. These results reflects the importance of improvement of pre-trained large language models and how these models enhancing the performance of different natural language processing tasks.
- Open Access
CommonLID: Re-evaluating State-of-the-Art Language Identification Performance on Web Data
Language identification (LID) is a fundamental step in curating multilingual corpora. However, LID models still perform poorly for many languages, especially on the noisy and heterogeneous web data often used to train multilingual language models. In this paper, we introduce CommonLID, a community-driven, human-annotated LID benchmark for the web domain, covering 109 languages. Many of the included languages have been previously under-served, making CommonLID a key resource for developing more representative high-quality text corpora. We show CommonLID's value by using it, alongside five other common evaluation sets, to test eight popular LID models. We analyse our results to situate our contribution and to provide an overview of the state of the art. In particular, we highlight that existing evaluations overestimate LID accuracy for many languages in the web domain. We make CommonLID and the code used to create it available under an open, permissive license.
Open AccessCommonLID: Re-evaluating State-of-the-Art Language Identification Performance on Web Data
Language identification (LID) is a fundamental step in curating multilingual corpora. However, LID models still perform poorly for many languages, especially on the noisy and heterogeneous web data often used to train multilingual language models. In this paper, we introduce CommonLID, a community-driven, human-annotated LID benchmark for the web domain, covering 109 languages. Many of the included languages have been previously under-served, making CommonLID a key resource for developing more representative high-quality text corpora. We show CommonLID's value by using it, alongside five other common evaluation sets, to test eight popular LID models. We analyse our results to situate our contribution and to provide an overview of the state of the art. In particular, we highlight that existing evaluations overestimate LID accuracy for many languages in the web domain. We make CommonLID and the code used to create it available under an open, permissive license.
- Open Access
- 1 January 2026
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Access
- Vol. 14, 139667-139682
An Interpretable Hybrid Ensemble Model for Paraphrase Detection With SHAP-Based Explanations
- 1 January 2026
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Access
- Vol. 14, 139667-139682
Open AccessAn Interpretable Hybrid Ensemble Model for Paraphrase Detection With SHAP-Based Explanations
- Open Access
- 1 January 2026
- Published by Association for Computational Linguistics (ACL)
- p. 493-496
REGLAT at AbjadGenEval: Multi-Model Ensemble Approach for Arabic AI-Generated Text Detection
Mariam Labib Francies, Nsrin Ashraf, Ahmed Megahed Fetouh, Hamada Nayel. Proceedings of the 2nd Workshop on NLP for Languages Using Arabic Script. 2026.
- 1 January 2026
- Published by Association for Computational Linguistics (ACL)
- p. 493-496
Open AccessREGLAT at AbjadGenEval: Multi-Model Ensemble Approach for Arabic AI-Generated Text Detection
Mariam Labib Francies, Nsrin Ashraf, Ahmed Megahed Fetouh, Hamada Nayel. Proceedings of the 2nd Workshop on NLP for Languages Using Arabic Script. 2026.
- Open Access
- 1 January 2026
- Published by Association for Computational Linguistics (ACL)
- p. 394-397
REGLAT at AbjadMed: Handling Imbalanced Arabic Medical Text Classification via Hierarchical KNN-MLP Architecture
Ahmed Megahed Fetouh, Mohammed Rahmath, Omer Dawood, Mariam Labib, Nsrin Ashraf, Hamada Nayel. Proceedings of the 2nd Workshop on NLP for Languages Using Arabic Script. 2026.
- 1 January 2026
- Published by Association for Computational Linguistics (ACL)
- p. 394-397
Open AccessREGLAT at AbjadMed: Handling Imbalanced Arabic Medical Text Classification via Hierarchical KNN-MLP Architecture
Ahmed Megahed Fetouh, Mohammed Rahmath, Omer Dawood, Mariam Labib, Nsrin Ashraf, Hamada Nayel. Proceedings of the 2nd Workshop on NLP for Languages Using Arabic Script. 2026.
- Open Access
- 11 December 2025
- Published by Elsevier in Data in Brief
- Vol. 64, 112381
DEAST: A dataset for English-Arabic scientific translation and vice versa
- 11 December 2025
- Published by Elsevier in Data in Brief
- Vol. 64, 112381
Open AccessDEAST: A dataset for English-Arabic scientific translation and vice versa
- 14 August 2025
- Published by Springer Nature in Neural Computing and Applications
- Vol. 37 (26) , 22029-22055
Swarm intelligence for handling out-of-vocabulary in Arabic Dialect Identification with different representations
- 14 August 2025
- Published by Springer Nature in Neural Computing and Applications
- Vol. 37 (26) , 22029-22055
Swarm intelligence for handling out-of-vocabulary in Arabic Dialect Identification with different representations