On-going Research

Research Papers (Accepted)

Title: Explainable AI for Stroke Risk Assessment: Voting versus Stacking Classifiers

Brief Description: This research focuses on the application of machine learning (ML) techniques for accurate stroke prediction, focusing on Voting and Stacking classifiers. Our study integrates feature significance analysis with Local Interpretable Model-agnostic Explanations (LIME) to enhance interpretability, thereby transcending traditional correctness measures.

 Journal: Journal of Artificial Intelligence (IJ-AI) [Q2 in AI Category]

Status: Accepted

Research Papers (Under Peer Review)

Title: Blended HCI-Based Usability Assessment with Convergence Metrics for e-Learning Platforms

Brief Description: We proposed a data-driven blended HCI framework while combining the users' and experts' perspectives to evaluate four popular Bangladeshi e-learning web platforms. This study integrates user experience ratings with expert heuristic evaluation to provide a balanced and holistic assessment of e-learning platform usability. The proposed convergence framework highlights where user perceptions align with or contradict expert judgments, while the composite usability index reveals Bohubrihi as the most user-centered and well-designed platform.

Journal: Computers in Human Behavior Reports [Q1 Journal]

Status: Peer Reviewing Stage

Title: CrosGrpsABS: Cross-Attention over Syntactic and Semantic Graphs for Aspect-Based Sentiment Analysis in a Low-Resource Language

Brief Description: This work addresses the underexplored task of aspect-based sentiment analysis (ABSA) in Bengali, a low-resource language lacking annotated datasets and NLP tools. We propose CrosGrpsABS, a hybrid framework that combines transformer-based contextual embeddings with syntactic and semantic graphs via a bidirectional cross-attention mechanism.

Journal: Expert Systems with Applications [Q1 Journal]

Status: Peer Reviewing Stage

Title: Hierarchical Graph Attention Networks with BERT Embedding for Bengali Aspect-Based Sentiment Analysis

Brief Description: This work addresses the challenges of Aspect-Based Sentiment Analysis (ABSA) in Bengali, a low-resource language, by introducing a novel hierarchical graph-based model. Leveraging Graph Attention Networks (GATs) and BERT embedding, complemented by a custom-designed Transformer block, our approach effectively captures the intricate relationships between aspect terms and emotion expressions.

Journal: Egyptian Informatics Journal [Q1 Journal]

Status: Peer Reviewing Stage

Title: Evaluating Assistive Apps for Visually Impaired Users: Exploration of App Features and Design Considerations

Brief Description: We propose a framework that will serve as a valuable resource for future app developers focusing on creating modified applications for visually impaired individuals. Our findings highlight the importance of accessibility features and user-centric design, suggesting future research should explore the integration of emerging technologies to further enhance app usability for visually impaired users.

 Journal: Multimedia Tools and Applications [Q1 Journal]

Status: Revision On-Going

Title: Enhancing Transparency in Healthcare: An Explainable AI Framework for Multi-Disease Diagnosis

Brief Description: This paper proposes an Explainable Artificial Intelligence (XAI) framework aimed at enhancing transparency in multi-disease diagnosis by integrating explainability into both tabular and image-based data modalities. This dual-modality framework effectively addresses the transparency gap in AI-powered diagnostics by combining performance with interpretability. Our results demonstrate that integrating multiple XAI techniques across ML and DL models significantly improves clinical relevance, fosters clinician trust, and paves the way for ethical and accountable AI systems in healthcare.

 Journal: Cognitive Computation [Q1 Journal]

Status: Peer Reviewing Stage

Title: Dynamic Span Interaction and Graph-Aware Memory for Entity-Level Sentiment Classification

Brief Description: SpanEIT is a span-based framework designed to tackle the complexities of entity-level sentiment classification, including subtle entity–sentiment interactions, cross-sentence dependencies, and coreference consistency. It models fine-grained span interactions with bidirectional attention, incorporates syntactic and co-occurrence relations through a graph attention network, and maintains entity coherence using a coreference-aware memory module. Experiments on FSAD, BARU, and IMDB demonstrate that SpanEIT surpasses state-of-the-art baselines in accuracy and F1, highlighting its effectiveness for real-world sentiment analysis tasks.

 Journal: IEEE Transactions on Audio, Speech and Language Processing [Q1 Journal]

Status: Peer Reviewing Stage

Title: AHCL: Capability-Aware Mode Adaptation for Privacy-Preserving Federated Learning

Brief Description: Adaptive Hybrid Collaborative Learning (AHCL) is a capability-aware decentralized framework that assigns clients to FL, SL, or GOA based on latency, energy, bandwidth, and data characteristics to overcome heterogeneity in healthcare settings. It uses a lightweight pairwise-masked secure aggregation protocol to prevent gradient and activation leakage without heavy cryptographic overhead. Experiments on three healthcare datasets show that AHCL improves accuracy and F1 by 3–8%, reduces time and energy use by 17–38%, and lowers inference-attack success rates to near-random levels, making it suitable for clinical and wearable environments.

 Journal: IEEE Transactions on Emerging Topics in Computational Intelligence [Q1 Journal]

Status: Peer Reviewing Stage

Research Papers (On-Going)

Title: 

Brief Description: 

 Journal: XYZ [Q1 Journal]

Status: On-going