Aliyu Umar
Current Organizations / Institutes
The organizations/institutes from the scholar's latest article
Previous Organizations / Institutes
The organizations / institutes from the scholar's other articles in the past ten years
Research Subjects
Aliyu Umar has a total of 21 co-authors. Here are the top co-authors by number of shared publications:
- Adamu Lawan
Beihang University
3 shared publications - Abdullahi Abdulkadir
University of Portsmouth
2 shared publications - Suleiman Farouk
Baze University
2 shared publications - Abdullahi Mohammed Ibrahim
Jigawa State Polytechnic Dutse
1 shared publication - Adamu Sani Yahaya
Department of Computer Science, COMSATS University Islamabad, Islamabad 44000, Pakistan
1 shared publication - Buhari Aliyu
Jigawa State Polytechnic Dutse
1 shared publication
Latest Publications
- 4 November 2026
- Published by Springer Nature
- p. 188-198
Detecting Dark Patterns in User Interfaces Using Logistic Regression and Bag-of-Words Representation
- 4 November 2026
- Published by Springer Nature
- p. 188-198
Detecting Dark Patterns in User Interfaces Using Logistic Regression and Bag-of-Words Representation
- 1 August 2026
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Transactions on Emerging Topics in Computational Intelligence
- Vol. 10 (4) , 2843-2855
DualKanbaFormer: An Efficient Selective Sparse Framework for Multimodal Aspect-Based Sentiment Analysis
Multimodal Aspect-Based Sentiment Analysis (MABSA) enhances sentiment detection by integrating text with complementary modalities, such as images, for a more comprehensive understanding of aspect-level sentiment. Despite notable progress, existing MABSA methods still lack strong aspect awareness, particularly in sentences with multiple aspects and conflicting sentiments or when visual cues are misaligned with the text. Together, these issues hinder accurate modeling of aspect-focused and cross-modal dependencies, limiting the effectiveness of MABSA. To address these challenges, we propose DualKanbaFormer, a novel framework with parallel Textual and Visual KanbaFormer modules tailored for MABSA. At its core, the model introduces Aspect-Driven Sparse Attention (ADSA) with an aspect-aware attention gating mechanism, which dynamically regulates the contributions of Scope, Focus, and Proximity attention branches conditioned on the target aspect. This ensures precise alignment between aspects and opinion signals. In addition, the selective state space model Mamba strengthens aspect–opinion alignment by modeling long-range dependencies across modalities. To enhance representational expressiveness and training stability, Kolmogorov–Arnold Networks (KANs) and Dynamic Tanh (DyT) replace conventional feed-forward and normalization layers. Finally, a CNN-based multimodal gated fusion module adaptively regulates cross-modal interactions, ensuring robust alignment between textual and visual features. Experiments on two benchmark datasets demonstrate that DualKanbaFormer outperforms several state-of-the-art models.
- 1 August 2026
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Transactions on Emerging Topics in Computational Intelligence
- Vol. 10 (4) , 2843-2855
DualKanbaFormer: An Efficient Selective Sparse Framework for Multimodal Aspect-Based Sentiment Analysis
Multimodal Aspect-Based Sentiment Analysis (MABSA) enhances sentiment detection by integrating text with complementary modalities, such as images, for a more comprehensive understanding of aspect-level sentiment. Despite notable progress, existing MABSA methods still lack strong aspect awareness, particularly in sentences with multiple aspects and conflicting sentiments or when visual cues are misaligned with the text. Together, these issues hinder accurate modeling of aspect-focused and cross-modal dependencies, limiting the effectiveness of MABSA. To address these challenges, we propose DualKanbaFormer, a novel framework with parallel Textual and Visual KanbaFormer modules tailored for MABSA. At its core, the model introduces Aspect-Driven Sparse Attention (ADSA) with an aspect-aware attention gating mechanism, which dynamically regulates the contributions of Scope, Focus, and Proximity attention branches conditioned on the target aspect. This ensures precise alignment between aspects and opinion signals. In addition, the selective state space model Mamba strengthens aspect–opinion alignment by modeling long-range dependencies across modalities. To enhance representational expressiveness and training stability, Kolmogorov–Arnold Networks (KANs) and Dynamic Tanh (DyT) replace conventional feed-forward and normalization layers. Finally, a CNN-based multimodal gated fusion module adaptively regulates cross-modal interactions, ensuring robust alignment between textual and visual features. Experiments on two benchmark datasets demonstrate that DualKanbaFormer outperforms several state-of-the-art models.
- Top Cited Papers
- 5 March 2026
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Internet of Things Journal
- Vol. 13 (10) , 20477-20497
Anomaly Detection in IoT Environment Using Machine Learning: A Survey
The explosive growth of the Internet of Things (IoT) has introduced vast amounts of data and unprecedented security challenges, making effective anomaly detection in IoT environments a critical concern. This article presents a comprehensive survey of machine learning techniques for IoT anomaly detection, with an emphasis on their applicability, performance tradeoffs, and alignment with real-world requirements. We adhere to a systematic literature review methodology (PRISMA) to identify and analyze recent advances, categorizing them into supervised, unsupervised, semi-supervised, and deep learning approaches, as well as emerging paradigms like federated learning (FL). For each category, we discuss representative algorithms and architectures, ranging from isolation forests and one-class SVMs (ocSVMs) to deep autoencoders and graph neural networks, and highlight how they address issues such as data imbalance, concept drift, and limited labels in IoT data. We also review prominent IoT security datasets and evaluation metrics used in the field and present illustrative results from experiments and simulations of the investigated techniques. Our comparative analysis underscores that while supervised and deep learning models achieve high accuracy under ample labeled data, unsupervised and hybrid methods offer robustness against novel attacks and evolving network behavior. Finally, we identify open challenges (e.g., adaptive model updating, feature selection for high-dimensional IoT data, and privacy-preserving detection) and outline future research directions toward resilient, real-time anomaly detection in large-scale IoT networks.
- 5 March 2026
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Internet of Things Journal
- Vol. 13 (10) , 20477-20497
Top Cited PapersAnomaly Detection in IoT Environment Using Machine Learning: A Survey
The explosive growth of the Internet of Things (IoT) has introduced vast amounts of data and unprecedented security challenges, making effective anomaly detection in IoT environments a critical concern. This article presents a comprehensive survey of machine learning techniques for IoT anomaly detection, with an emphasis on their applicability, performance tradeoffs, and alignment with real-world requirements. We adhere to a systematic literature review methodology (PRISMA) to identify and analyze recent advances, categorizing them into supervised, unsupervised, semi-supervised, and deep learning approaches, as well as emerging paradigms like federated learning (FL). For each category, we discuss representative algorithms and architectures, ranging from isolation forests and one-class SVMs (ocSVMs) to deep autoencoders and graph neural networks, and highlight how they address issues such as data imbalance, concept drift, and limited labels in IoT data. We also review prominent IoT security datasets and evaluation metrics used in the field and present illustrative results from experiments and simulations of the investigated techniques. Our comparative analysis underscores that while supervised and deep learning models achieve high accuracy under ample labeled data, unsupervised and hybrid methods offer robustness against novel attacks and evolving network behavior. Finally, we identify open challenges (e.g., adaptive model updating, feature selection for high-dimensional IoT data, and privacy-preserving detection) and outline future research directions toward resilient, real-time anomaly detection in large-scale IoT networks.
- Open Access
- 18 June 2025
- Published by Science Publishing Group in International Journal of Information and Communication Sciences
- Vol. 10 (2) , 34-40
Unveiling the Dark Web: Enhancing Machine Learning Models for Deciphering Illicit Communication Patterns
The dark web, an obscured and encrypted segment of the internet, serves as a nexus for illicit activities and underground communities, presenting substantial challenges for law enforcement and cybersecurity professionals. This paper explains the linguistic patterns and communication dynamics within the dark web by using the Random Forest classifier model. The Random Forest Classifier, selected for its proficiency in managing high-dimensional and noisy data, is utilized to classify and decode the cryptic language prevalent on the dark web. The model demonstrates a high accuracy of 98%, complemented by strong precision, recall, and F1-score metrics. These findings underscore the model's efficacy in identifying significant linguistic patterns and offer valuable insights into the communication mechanisms within dark web communities. Despite the promising results, this study acknowledges data quality and generalizability limitations, proposing avenues for future research to enhance model robustness and address ethical considerations in dark web analytics. This work contributes to the ongoing efforts to understand and mitigate illicit activities on the dark web through the application of machine learning and linguistic analysis.
- 18 June 2025
- Published by Science Publishing Group in International Journal of Information and Communication Sciences
- Vol. 10 (2) , 34-40
Open AccessUnveiling the Dark Web: Enhancing Machine Learning Models for Deciphering Illicit Communication Patterns
The dark web, an obscured and encrypted segment of the internet, serves as a nexus for illicit activities and underground communities, presenting substantial challenges for law enforcement and cybersecurity professionals. This paper explains the linguistic patterns and communication dynamics within the dark web by using the Random Forest classifier model. The Random Forest Classifier, selected for its proficiency in managing high-dimensional and noisy data, is utilized to classify and decode the cryptic language prevalent on the dark web. The model demonstrates a high accuracy of 98%, complemented by strong precision, recall, and F1-score metrics. These findings underscore the model's efficacy in identifying significant linguistic patterns and offer valuable insights into the communication mechanisms within dark web communities. Despite the promising results, this study acknowledges data quality and generalizability limitations, proposing avenues for future research to enhance model robustness and address ethical considerations in dark web analytics. This work contributes to the ongoing efforts to understand and mitigate illicit activities on the dark web through the application of machine learning and linguistic analysis.
- 31 May 2020
- Published by Doorma Journals in International Journal of Computer Networks and Communications Security
- Vol. 8 (5) , 37-45
An Efficient Approach of Threat Hunting Using Memory Forensics
The capacity and occurrence of new cyber-attacks have shattered in recent years. Such measures have very complicated workflows and comprise multiple illegal actors and organizations. Threat hunting demonstrates the process of proactively searching through networks for threats based on zero-day attacks by repeating the hunting process again and again. Unlike threat intelligence, it uses different automated security tools to collect logs in order to provide a pattern for making new intelligence-based tools by following those logs. According to our research findings about “threat hunting tools” there’s a major flaw that the designed tools are limited to the collection of logs. It works completely on logs for generating new patterns avoiding system’s main memory. Codes written directly to memory fail this process to provide proactive hunting. To overcome this major challenge, we are proposing two distinct methods, either by generating malicious code alerts or by binding memory forensics processes with threat hunting tools to make active hunting possible
- 31 May 2020
- Published by Doorma Journals in International Journal of Computer Networks and Communications Security
- Vol. 8 (5) , 37-45
An Efficient Approach of Threat Hunting Using Memory Forensics
The capacity and occurrence of new cyber-attacks have shattered in recent years. Such measures have very complicated workflows and comprise multiple illegal actors and organizations. Threat hunting demonstrates the process of proactively searching through networks for threats based on zero-day attacks by repeating the hunting process again and again. Unlike threat intelligence, it uses different automated security tools to collect logs in order to provide a pattern for making new intelligence-based tools by following those logs. According to our research findings about “threat hunting tools” there’s a major flaw that the designed tools are limited to the collection of logs. It works completely on logs for generating new patterns avoiding system’s main memory. Codes written directly to memory fail this process to provide proactive hunting. To overcome this major challenge, we are proposing two distinct methods, either by generating malicious code alerts or by binding memory forensics processes with threat hunting tools to make active hunting possible