Muawia A. Elsadig

Dr. Muawia A. Elsadig

orcid

Research Focus & Metrics

Academic Career Length

11Years (2016-Present)

(Calculated once Scilit publication data is synced: current year − year of first publication + 1)

Representative Publications

View All Publications
16 February 2024 in IEEE Access
Verified
Journal Article
Open access
YAYasir Abdugadir
Yasir Abdugadir
AKAkbar Khanan
Akbar Khanan
MBMohamed Bashir
Mohamed Bashir
AMAbdul Hakim H M Mohamed
Abdul Hakim H M Mohamed
Mousab Alata Elseddig AdielMousab Alata Elseddig Adiel
Mousab Alata Elseddig Adiel
Muawia A. ElsadigMuawia A. Elsadig
Muawia A. Elsadig
In the context of a more linked and globalized society, the significance of proficient cross-cultural communication has been increasing to a position of utmost importance. Language functions as a crucial medium that establishes connections among people, corporations, and countries, thus demanding the implementation of precise and effective translation systems. This comprehensive review paper aims to contribute to the evolving landscape of AI-driven language translation by critically examining the existing literature, identifying key debates, and uncovering areas of innovation and limitations where the primary objective -is to provide a nuanced understanding of the current state of AI-driven language translation, along with emphasizing the advancements, challenges, and ethical considerations. In this review, ongoing debates surrounding AI-driven language translations were actively involved. By evaluating different viewpoints and methodologies, insights into unresolved questions that contribute to broader discourse in the field were provided. The future trajectory of this study is the incorporation of cross-lingual dialect adaptability and the advancement of Artificial Intelligence translation systems, with a focus on prioritizing inclusion and cultural understanding.

7 August 2023 in IEEE Access
Verified
Journal Article
Open access
Muawia A. ElsadigMuawia A. Elsadig
Muawia A. Elsadig
The characteristics and performance of wireless sensor networks (WSNs) are the main reasons for their rapid expansion in various fields. However, these networks are extremely susceptible to multiple security assaults, including denial-of-service (DoS) attacks, which are among the most prevalent in these networks. This study sheds light on WSN restrictions, weaknesses, and security threats with a focus on DoS attacks. Recent techniques for DoS attack detection have been investigated thoroughly, highlighting their achievements and limitations. This provides valuable insight into the current state of recent research in this field. Accordingly, this study proposes a lightweight machine learning detection approach based on a decision tree (DT) algorithm with the Gini feature selection method to detect DoS attacks in WSNs. An enhanced version of the WSN-DS dataset, developed by the author, was used to train and test the proposed approach. The proposed approach has shown good performance by achieving an accuracy rate of 99.5% with minimum overhead compared to random forest (RF), extreme gradient boosting (XGBoost), and k-nearest neighbor (KNN) classifiers. It only takes 9.7%, 13%, and 2% of the processing time required by FR, XGBoost, and KNN respectively, which indicates that our proposed approach significantly outperforms these classifiers in terms of processing time. It is noteworthy that RF achieved an accuracy that was somewhat superior; however, the proposed approach greatly surpassed RF by taking only 9.7% of the RF processing time, which is an important factor in meeting WSN constraints.

4 April 2022 in IEEE Access
Verified
Journal Article
Open access
Muawia A. ElsadigMuawia A. Elsadig
Muawia A. Elsadig
AGAhmed Gafar
Ahmed Gafar
The advanced development of computer networks and communication technologies has made covert communications easier to construct, faster, undetectable and more secure than ever. A covert channel is a path through which secret messages can be leaked by violating a system security policy. The detection of such dangerous, unwatchable, and hidden threats is still one of the most challenging aspects. This threat exploits methods that are not dedicated to communication purposes, meaning that traditional security measures fail to detect its existence. This review has introduced a brief introduction of covert channel definitions, types and developments, with a particular focus on detection techniques using machine learning (ML) approaches. It provides a thorough review of the most common covert channels and ML techniques that are used to counter them, as well as addressing their achievements and limitations. In addition, this paper introduces a comparative experimental study for some common ML approaches that are commonly used in this field. Accordingly, the performance of these classifiers was evaluated and reported. The paper concludes that our information is still at risk, nothing is said to be secured and more work on the detection of covert channels is required.

1 February 2023 in International Journal of Electrical and Computer Engineering (IJECE)
Verified
Journal Article
Open access
PDF file
Muawia A. ElsadigMuawia A. Elsadig
Muawia A. Elsadig
AAAbdelrahman Altigani
Abdelrahman Altigani
HEHuwaida Tagelsir Elshoush
Huwaida Tagelsir Elshoush
As the cause of the breast cancer disease has not yet clearly identified and a method to prevent its occurrence has not yet been developed, its early detection has a significant role in enhancing survival rate. In fact, artificial intelligent approaches have been playing an important role to enhance the diagnosis process of breast cancer. This work has selected eight classification models that are mostly used to predict breast cancer to be under investigation. These classifiers include single and ensemble classifiers. A trusted dataset has been enhanced by applying five different feature selection methods to pick up only weighted features and to neglect others. Accordingly, a dataset of only 17 features has been developed. Based on our experimental work, three classifiers, multi-layer perceptron (MLP), support vector machine (SVM) and stack are competing with each other by attaining high classification accuracy compared to others. However, SVM is ranked on the top by obtaining an accuracy of 97.7% with classification errors of 0.029 false negative (FN) and 0.019 false positive (FP). Therefore, it is noteworthy to mention that SVM is the best classifier and it outperforms even the stack classier.

13 January 2021 in IEEE Access
Verified
Journal Article
Open access
AAAbdelrahman Altigani
Abdelrahman Altigani
SHShafaatunnur Hasan
Shafaatunnur Hasan
BBBazara Barry
Bazara Barry
SNShiraz Naserelden
Shiraz Naserelden
Muawia A. ElsadigMuawia A. Elsadig
Muawia A. Elsadig
HEHuwaida T. Elshoush
Huwaida T. Elshoush
To secure various forms of data, a polymorphic variant of the Advanced Encryption Standard (P-AES) has been introduced. In the P-AES, the AES parameters' values will change with every new key. The exact values will be available only to legitimate communicating parties during execution time. To achieve these objectives, the basic AES transformations, SubBytes, ShiftRows, and MixColumns, have been made key-dependent in the proposed P-AES. Hence, with every new key, these transformations will operate differently. The receiver can retrieve the operations' details from the encryption key. Consequently, polymorphism has been achieved and interoperability remains intact. P-AES has been implemented seamlessly using the existing AES modules, and the performance was more or less equal to the AES performance (71 and 70 milliseconds to encrypt 500 bytes using the P-AES and the AES respectively). From a security standpoint, the proposed P-AES fully complies with Kerckhoff's principle. This means the cipher has an open design, and the security provided by the P-AES depends only on the secrecy of the encryption key. The cipher resistance to differential and linear attacks has been proved. Moreover, the resulting proposed cipher can operate in 128 different ways, which will significantly reduce the capabilities of any sophisticated attacker. Furthermore, the proposed P-AES's scores of the key avalanche and the plaintext avalanche were 0.496 and 0.504 respectively. Finally, the Statistical Test Suite (STS) recommended by the NIST has been used to ensure the randomness of the cipher output, and the cipher has passed all the STS tests.

25 December 2024 in IEEE Access
Verified
Journal Article
Open access
YAYasir Abdugadir
Yasir Abdugadir
AMAbdul Hakim H M Mohamed
Abdul Hakim H M Mohamed
AKAkbar Khanan
Akbar Khanan
MBMohamed Bashir
Mohamed Bashir
Mousab Alata Elseddig AdielMousab Alata Elseddig Adiel
Mousab Alata Elseddig Adiel
Muawia A. ElsadigMuawia A. Elsadig
Muawia A. Elsadig
This review examines the ethical, social, and technical challenges posed by AI-generated text tools, focusing on their rapid advancement and widespread adoption. An exhaustive literature search across many databases, strict inclusion/exclusion criteria, and a rigorous analysis procedure are all parts of our systematic review technique. This guarantees an impartial and complete study of the current status of AI-generated text tools. The study analyzes prominent language models, including GPT-3, GPT-4, LaMDA, PaLM, Claude, Jasper, and Llama 2, evaluating their capabilities in natural language processing and generation. The analysis reveals significant advancements, with GPT-3 demonstrating a 92% accuracy rate on standard natural language understanding benchmarks, outperforming LaMDA (88%) and PaLM (85%). To illustrate real-world implications, the review presents a case study of ChatGPT's application in healthcare, where it achieved 80% consistency with expert opinions in assessing acute ulcerative colitis. This case highlights both the potential benefits and ethical concerns of AI in critical domains. Quantitative bias analysis shows that GPT-3 generated biased content in 15% of test cases involving sensitive topics, a higher rate than LaMDA (12%) and PaLM (10%). We provide an in-depth analysis of fairness and bias issues, particularly in image generation tasks depicting professional roles. Our research synthesizes insights from technical advancements, ethical considerations, and real-world applications across healthcare, education, and creative sectors. We address critical privacy concerns and data protection challenges, noting struggles in AI-generated text detection and investigating AI's potential in enabling cyberattacks. We underscore the need for comprehensive governance systems and multidisciplinary cooperation. To provide a cohesive analysis of the ethical considerations surrounding AI-generated text tools, we employ a multifaceted ethical framework drawing on established theories. Utilitarianism, which seeks to maximize happiness for everyone; deontology, which places an emphasis on right and wrong; and Virtue Ethics, which analyzes the moral nature of deeds and actors, are all included in this framework. In this article, we use this approach to investigate AI ethics from a variety of angles, including privacy, prejudice, and social implications, as well as concerns of justice and fairness. Moreover, the study critically examines existing and proposed legal frameworks addressing AI ethics, identifying regulatory gaps and proposing adaptive policy recommendations to address the unique challenges posed by AI-generated text tools. Our review contributes a critical analysis of AI-generated text tools, their impacts, and the need for responsible innovation. The study provides precise guidelines for the ethical development and implementation of AI, highlighting the need to strike a balance between technical progress and ethical concerns to guarantee that AI technologies have a beneficial effect on society while protecting human values. The emergence of generative artificial intelligence (AI) signifies a substantial revolution in our methods of interacting with language and information.