Shamsu Abdullahi

Shamsu Abdullahi

Auto-generated profile, based on publicly available information

Shamsu Abdullahi

Department of Computing, Faculty of Management, Science, and Computing, Universiti Teknologi PETRONAS, Perak, Malaysia

Research Focus & Metrics

Academic Career Length

2Years (2025-Present)

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

Representative Publications

View All Publications
28 January 2025 in IEEE Access
Verified
Journal Article
Open access
SAShamsu Abdullahi
Shamsu Abdullahi
KDKamaluddeen Usman Danyaro
Kamaluddeen Usman Danyaro
AZAbubakar Zakari
Abubakar Zakari
IAIzzatdin Abdul Aziz
Izzatdin Abdul Aziz
Noor Amila ZawawiNoor Amila Zawawi
Noor Amila Zawawi
SAShamsuddeen Adamu
Shamsuddeen Adamu
Large Language Models (LLMs) have transformed Natural Language Processing (NLP) and Software Engineering by fostering innovation, streamlining processes, and enabling data-driven decision-making. Recently, the adoption of LLMs in time-series analysis has catalyzed the emergence of time-series LLMs, a rapidly evolving research area. Existing reviews provide foundational insights into time-series LLMs but lack a comprehensive examination of recent advancements and do not adequately address critical challenges in this domain. This Systematic Literature Review (SLR) bridges these gaps by analysing state-of-the-art contributions in time-series LLMs, focusing on architectural innovations, tokenisation strategies, tasks, datasets, evaluation metrics, and unresolved challenges. Using a rigorous methodology based on PRISMA guidelines, over 700 studies from 2020 to 2024 were reviewed, with 59 relevant studies selected from journals, conferences, and workshops. Key findings reveal advancements in architectures and novel tokenization strategies tailored for temporal data. Forecasting dominates the identified tasks with 79.66% of the selected studies, while classification and anomaly detection remain underexplored. Furthermore, the analysis reveals a strong reliance on datasets from energy and transportation domains, highlighting the need for more diverse datasets. Despite these advancements, significant challenges persist, including tokenization inefficiencies, prediction hallucinations, and difficulties in modelling long-term dependencies. These issues hinder the robustness, scalability, and adaptability of time-series LLMs across diverse applications. To address these challenges, this SLR outlines a research roadmap emphasizing the improvement of tokenization methods, the development of mechanisms for capturing long-term dependencies, the mitigation of hallucination effects, and the design of scalable, interpretable models for diverse time-series tasks.

12 November 2025 in IEEE Access
Verified
Journal Article
Open access
UMUmama
Umama
KDKamaluddeen Usman Danyaro
Kamaluddeen Usman Danyaro
MNMAGED NASSER
MAGED NASSER
AZAbubakar Zakari
Abubakar Zakari
SAShamsu Abdullahi
Shamsu Abdullahi
SSSara Shoaib
Sara Shoaib
The rapid emergence of Large Language Models (LLMs) has significantly advanced the field of code generation, sparking growing research interest across both academia and industry. While existing reviews provide foundational insights into LLM-based code generation, they do not provide a comprehensive analysis of recent developments or adequately address the dual perspective of technical and demographic insights. This Systematic Literature Review (SLR) bridges critical gap in existing work by systematically analysing both technical and demographic dimensions of LLM-based code generation. Following a PRISMA-guided methodology 58 studies published between 2020 and 2025 were selected from four major digital libraries. Key findings reveal a sharp rise in evaluation studies and frameworkdriven contributions. Notably, HumanEval, MBPP and APPS emerged as dominant benchmarks. While functional-correctness metrics particularly pass@k are the most frequently adopted metrics. Moreover, the review uncovers a previously unexamined demographic landscape, highlighting, geographic distribution, institutional collaboration, team size, publishing trends, venue types, and the top 10 highly cited studies from the selected studies. Significantly, a high proportion of selected studies were published in conferences. Several persistent challenges found in existing reviews that hinder the reliability and applicability of LLM-based code generation. These include hallucinations, limited generalizability, security vulnerabilities, and a lack of interpretability. These limitations constrain the robustness, trustworthiness, and scalability of current systems. To address these issues, this reviewoffers actionable insights and strategic guidelines to guide future research, evaluation strategies, and practical implementations in LLM-based code generation.

12 December 2025
Verified
Book Chapter
Closed access
TCTahmid Ashraf Chowdhury
Tahmid Ashraf Chowdhury
FUFasee Ullah
Fasee Ullah
MSMujeeb Ur Rehman Shaikh
Mujeeb Ur Rehman Shaikh
SAShamsu Abdullahi
Shamsu Abdullahi
SHShoaib Ul Hassan
Shoaib Ul Hassan
MAMohibul Alam
Mohibul Alam
This chapter shows how blockchain improves transparency, efficiency, and trust in oil and gas. Supply chains and trade finance are slowed by fragmented systems, manual handoffs, delays, and fraud risk. A decentralized, tamper-evident ledger enables shared real-time data, end-to-end traceability, and secure records across operators, suppliers, shippers, banks, and regulators. Smart contracts automate terms, cut intermediaries, speed settlements, and improve liquidity. We map use cases in procurement, logistics tracking, asset integrity, and financial clearing, and pair blockchain with IoT telemetry for live monitoring and tokenization of commodities for digital trade. We also address regulatory uncertainty, scalability, and interoperability, and propose a stepwise framework for end-to-end rollout aligned to the oil and gas value chain. Uniting technology, operations, and governance, the chapter positions blockchain as a pillar of Industry 4.0 in energy, enabling secure, transparent, efficient global commerce.

17 March 2026 in IEEE Access
Verified
Journal Article
Open access
SAShamsu Abdullahi
Shamsu Abdullahi
KDKamaluddeen Usman Danyaro
Kamaluddeen Usman Danyaro
HCHaruna Chiroma
Haruna Chiroma
MYMuhammad Muntasir Yakubu
Muhammad Muntasir Yakubu
MYMuhammad Sabo Yahaya
Muhammad Sabo Yahaya
MZMusa Ahmed Zayyad
Musa Ahmed Zayyad
Tokenizing continuous time-series signals for Large Language Models (LLMs) remains a fundamental challenge due to the mismatch between numerical temporal data and discrete token-based architectures. Existing approaches (such as patching, symbolic aggregation, or quantization) address this problem in isolation, often sacrificing numerical fidelity, temporal coherence, or computational efficiency. We propose Hierarchical Symbolic–Quantized Patching (HSQP), a unified tokenization framework that hierarchically integrates patch segmentation, adaptive Brownian Bridge–based symbolic aggregation (ABBA), and affine quantization into a jointly embedded, non-separable token representation. Unlike sequential preprocessing pipelines, HSQP fuses symbolic cluster identities and quantized numerical descriptors into compact dual-semantic tokens that simultaneously encode structural temporal patterns and bounded numerical precision. The hierarchical design reduces sequence length, constrains reconstruction drift through patch-level anchoring, and introduces bounded quantization noise that stabilizes downstream Transformer attention. We provide theoretical analysis of quantization error propagation and demonstrate that reconstruction error remains locally bounded under Lipschitz continuity assumptions. Extensive experiments on six benchmark datasets show that HSQP consistently improves reconstruction fidelity and downstream forecasting accuracy while achieving stable compression ratios and balanced token entropy. Integrated as a plug-and-play module with frozen LLM backbones, HSQP enhances predictive performance without architectural modification. These results establish HSQP as an efficient, interpretable, and scalable tokenization paradigm for adapting continuous time-series data to LLM-based forecasting frameworks.

12 December 2025
Verified
Book Chapter
Closed access
SAShamsu Abdullahi
Shamsu Abdullahi
TATahmid Chowdhury Ashraf
Tahmid Chowdhury Ashraf
KDKamaluddeen Usman Danyaro
Kamaluddeen Usman Danyaro
FUFasee Ullah
Fasee Ullah
MYMuhammad Sabo Yahaya
Muhammad Sabo Yahaya
MSMujeeb Ur Rehman Shaikh
Mujeeb Ur Rehman Shaikh
Growing operational complexity, ageing infrastructure, and rising demands for efficiency and sustainability are all contributing factors to the oil and gas industry's current dramatic transformation. To improve operations in offshore facilities, this chapter investigates the combination of two innovative digital technologies: blockchain and large language models (LLMs). LLMs offer sophisticated features for handling unstructured technical data, automating documentation, enhancing decision-making, and assisting with predictive maintenance. Blockchain guarantees trust, transparency, and safe, decentralised data management throughout the value chain. Applications like smart contract automation, secure data exchange, real-time compliance, and intelligent asset monitoring are made possible by these technologies working together. The chapter also discusses issues with workforce development, data quality assurance, integration, and compatibility with legacy systems.

3 February 2026
Verified
Conference Paper
Open access
AAAl-Amin Abdulhamid
Al-Amin Abdulhamid
KKKhairul Shafee Kalid
Khairul Shafee Kalid
SSSavita K Sugathan
Savita K Sugathan
SAShamsu Abdullahi
Shamsu Abdullahi
MYMuhammad Sabo Yahaya
Muhammad Sabo Yahaya
SISarah Aisyah Isnani
Sarah Aisyah Isnani