MIMonzurul Islam

Monzurul Islam

*This is an algorithmically generated author record based on publicly available information.

Current Organizations / Institutes

The organizations/institutes from the scholar's latest article


Monzurul Islam has a total of 30 co-authors. Here are the top co-authors by number of shared publications:


Latest Publications

  • Crash Risk Analysis of Non-Motorists using Interpretable Tabular Deep Learning

    by Md Monzurul Islam, Anannya Ghosh Tusti, Mahmuda Sultana Mimi, Shriyank Somvanshi, Subasish Das

    The study investigates injury severity patterns among non-motorists (e.g., pedestrians and bicyclists) in Illinois using interpretable machine learning models applied to crash data from 2021 to 2023. The dataset included variables such as roadway design, environmental context, crash characteristics, and driver behavior. To address class imbalance in injury severity labels, synthetic oversampling was employed. Feature selection was performed using Random Forest and XGBoost, followed by training TabNet and FT-Transformer models. TabNet achieved the highest accuracy under the balanced training configuration and showed the strongest performance on the original imbalanced dataset across precision, recall, and F1 score, while also providing intrinsic interpretability through sequential attention-based feature selection. SHAP analysis revealed that temporal and environmental factors significantly influenced early model decisions, with behavioral and infrastructure-related variables impacting subsequent stages. The findings emphasize optimizing lighting, enhancing traffic control, and addressing time-specific crash risks. The study underscores the potential of interpretable tabular deep learning in traffic safety analytics, advocating for integration with contextual, geospatial, and behavioral data for targeted interventions.

  • Open Access

    Understanding Public Risk Perceptions of Alcohol- and Marijuana-Impaired Driving: A Structural Equation Modeling Approach Using Nationally Representative Survey Data

    by Sawgat Ahmed Shuvo, Sharif Ahmed Rafat, Md Monzurul Islam, Subasish Das

    Public perceptions of impaired driving influence support for traffic safety interventions and may be shaped by both individual characteristics and everyday driving experiences. This study investigates how driving exposure, roadway experiences, and sociodemographic factors influence public perceptions of alcohol- and marijuana-impaired driving as community safety concerns using nationally representative survey data from the Pew Research Center American Trends Panel Wave 152 (N = 5,410 U.S. adults; August 2024). Chi-square analyses identified significant associations between impaired-driving risk perception and age, gender, education, household income, and political ideology, with alcohol-impaired driving consistently perceived as a greater community safety problem than marijuana-impaired driving. Structural equation modeling (SEM) revealed that more frequent driving was associated with greater exposure to road rage, which contributed to perceptions of deteriorating roadway safety and heightened concern about impaired driving. Mediation analysis further demonstrated that road rage exposure partially mediated the relationship between driving frequency and impaired-driving risk perception, highlighting aggressive traffic environments as an important behavioral mechanism linking everyday driving experiences to public safety perceptions. These findings extend understanding of impaired-driving risk perception beyond demographic characteristics by demonstrating how routine roadway interactions shape public concern. The results contribute to traffic psychology by identifying behavioral pathways through which driving experiences influence safety perceptions and provide evidence to support behaviorally informed risk communication, public education initiatives, and enforcement strategies addressing both impaired and aggressive driving.

  • Classifying crash severity by SAE automation level using transformer models: Insights for automated vehicle policy

    by Mahmuda Sultana Mimi, Monzurul Islam, Swastika Barua, Tausif Islam Chowdhury, Anannya GhoshTusti, Gaurab Chhetri, Subasish Das


  • Quantum Computing in Transportation Engineering: A Survey

    by Shriyank Somvanshi, Subasish Das, Md Monzurul Islam, Sazzad Bin Bashar Polock, Gaurab Chhetri, Darrell Anderson

    Quantum computing introduces powerful computational paradigms capable of addressing optimization and machine learning challenges that are increasingly critical in transportation engineering. This paper presents a comprehensive survey of how quantum algorithms and models are being applied across transportation domains, including intelligent transportation systems, traffic management, vehicle routing, logistics, and autonomous systems. Unlike prior reviews that primarily introduce quantum theory, we systematically connect fundamental quantum principles-qubits, superposition, entanglement, and quantum gates-with specific transportation use cases. We synthesize recent findings showing that the Quantum Approximate Optimization Algorithm, Grover’s search, and hybrid quantum–classical solvers can improve performance in combinatorial problems such as vehicle routing and dynamic traffic signal control, while quantum machine learning models (e.g., Quantum Convolutional Neural Networks, Quantum Generative Adversarial Networks, and clustering algorithms) demonstrate promise in traffic prediction, anomaly detection, and autonomous vehicle perception. Our review highlights that although quantum hardware limitations (noise, scalability, connectivity) remain significant, empirical studies already suggest measurable improvements in solution quality and computational efficiency when compared to classical heuristics. The key contribution of this survey is to map the landscape of quantum computing in transportation engineering, identify application-specific advantages and current bottlenecks, and outline future research directions toward scalable, domain-specific quantum algorithms. By doing so, we provide transportation researchers and practitioners with a structured foundation to integrate quantum approaches into emerging mobility systems, with implications for efficiency, safety, sustainability, and equity.


  • Top Cited Papers
    Open Access

    Bridging the Black Box: A Survey on Mechanistic Interpretability in AI

    by Shriyank Somvanshi, Monzurul Islam, Amir Rafe, Anannya Ghosh Tusti, Arka Chakraborty, Anika Baitullah, Tausif Islam Chowdhury, Nawaf Alnawmasi, Anandi Dutta, Subasish Das

    Mechanistic interpretability seeks to reverse-engineer the internal logic of neural networks by uncovering human-understandable circuits, algorithms, and causal structures that drive model behavior. Unlike post hoc explanations that describe what models do, this paradigm focuses on why and how they compute, tracing information flow through neurons, attention heads, and activation pathways. This survey provides a high-level synthesis of the field-highlighting its motivation, conceptual foundations, and methodological taxonomy rather than enumerating individual techniques. We organize mechanistic interpretability across three abstraction layers— neurons , circuits , and algorithms —and three evaluation perspectives: behavioral , counterfactual , and causal . We further discuss representative approaches and toolchains that enable structural analysis of modern AI systems, outlining how mechanistic interpretability bridges theoretical insights with practical transparency. Despite rapid progress, challenges persist in scaling these analyses to frontier models, resolving polysemantic representations, and establishing standardized causal benchmarks. By connecting historical evolution, current methodologies, and emerging research directions, this survey aims to provide an integrative framework for understanding how mechanistic interpretability can support transparency, reliability, and governance in large-scale AI.


  • Top Cited Papers
    Open Access

    From Tiny Machine Learning to Tiny Deep Learning: A Survey

    by Shriyank Somvanshi, Monzurul Islam, Gaurab Chhetri, Rohit Chakraborty, Mahmuda Sultana Mimi, Sawgat Ahmed Shuvo, Kazi Sifatul Islam, Syed Javed, Sharif Ahmed Rafat, Anandi Dutta, et al.

    The rapid growth of edge devices has driven the demand for deploying artificial intelligence (AI) at the edge, giving rise to Tiny Machine Learning (TinyML) and its evolving counterpart, Tiny Deep Learning (TinyDL). While TinyML initially focused on enabling simple inference tasks on microcontrollers, the emergence of TinyDL marks a paradigm shift toward deploying deep learning models on severely resource-constrained hardware. This survey presents a comprehensive overview of the transition from TinyML to TinyDL, encompassing architectural innovations, hardware platforms, model optimization techniques, and software toolchains. We analyze state-of-the-art methods in quantization, pruning, and neural architecture search (NAS), and examine hardware trends from MCUs to dedicated neural accelerators. Furthermore, we categorize software deployment frameworks, compilers, and AutoML tools enabling practical on-device learning. Applications across domains such as computer vision, audio recognition, healthcare, and industrial monitoring are reviewed to illustrate the real-world impact of TinyDL. Finally, we identify emerging directions including neuromorphic computing, federated TinyDL, edge-native foundation models, and domain-specific co-design approaches. This survey aims to serve as a foundational resource for researchers and practitioners, offering a holistic view of the ecosystem and laying the groundwork for future advancements in edge AI.


  • Open Access

    ST-GraphNet: A Spatio-Temporal Graph Neural Network for Understanding and Predicting Automated Vehicle Crash Severity

    by Mahmuda Sultana Mimi, Monzurul Islam, Anannya Ghosh Tusti, Shriyank Somvanshi, Subasish Das, Xinyue Ye