Istiaque Bhuiyan

Mr. Istiaque Bhuiyan

orcid

Research Focus & Metrics

Academic Career Length

1Year (2026-Present)

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

Representative Publications

View All Publications
8 June 2026 in FinTech
Verified
Journal Article
Open access
PDF file
IBIstiaque Bhuiyan
Istiaque Bhuiyan
Muhammad Haseeb AhmadMuhammad Haseeb Ahmad
Muhammad Haseeb Ahmad
AHAriful Hoque
Ariful Hoque
Tanvir BhuiyanTanvir Bhuiyan
Tanvir Bhuiyan
This study examines customer retention intention in neobanking environments using a theory-informed explainable machine learning framework. Existing digital banking research typically relies on linear modelling approaches to explain retention behaviour, potentially overlooking nonlinear, value-range-dependent, and interaction-based predictive patterns. Using a publicly available survey of 305 neobank users, this study compares regularized linear models, a partial least squares structural equation modelling (PLS-SEM)-inspired benchmark, and XGBoost (version 3.2.0) under repeated nested cross-validation. SHapley Additive exPlanations (SHAP)-based explainability, SHAP interaction analysis, generalized additive model (GAM) diagnostics, construct-level aggregation, and construct-sensitivity checks are used to interpret model behaviour and assess robustness. The results show that XGBoost substantially outperforms the linear benchmarks, achieving the lowest average RMSE and highest average R2 across 100 out-of-sample test-fold estimates. Trust-related indicators provide the largest share of model-based predictive importance, followed by perceived security and switching costs. SHAP and GAM diagnostics suggest that trust and switching costs may contribute to retention intention in heterogeneous and nonlinear ways, while perceived security displays a more stable positive predictive pattern. Age-related nonlinearities appear weak and should be interpreted cautiously given the young sample profile. The analysis also suggests possible non-additive relationships between trust and perceived security. The study contributes to digital banking and FinTech research by showing how explainable machine learning can complement theory-driven retention models, identify potentially nonlinear predictive patterns, and preserve interpretability. The findings offer practical insight for trust-building, visible security assurance, and retention diagnostics in neobanking contexts.

14 May 2026 in Preprints.org
Verified
Preprint
Open access
PDF file
IBIstiaque Bhuiyan
Istiaque Bhuiyan
Tanvir BhuiyanTanvir Bhuiyan
Tanvir Bhuiyan
Purpose: This study examines whether finance-adapted (FA) phishing detection models improve detection of finance-themed (FT) attacks, whether improvements differ across email and webpage modalities, and whether finance adaptation creates a specialisation–generalisation trade-off. Design/Methodology/Approach: A domain-aware framework is developed using email (164,972 instances) and webpage (11,430 instances) datasets. FT and non-finance-themed (NFT) instances are identified using weighted lexicon-based labelling. Generic models are compared with FA models across Logistic Regression, Linear SVC, and Random Forest using F1-score, MCC, balanced accuracy, ROC-AUC, and PR-AUC. Statistical validation employs bootstrap confidence intervals and McNemar's test, while SHAP and permutation importance interpret webpage model behaviour. Findings: FA models outperform generic models in FT email classification, confirming that finance-specific semantic cues improve detection. However, gains are weaker and less consistent in webpage classification, where models rely mainly on structural indicators (page rank, Google index, hyperlinks). Results reveal a specialisation–generalisation trade-off: FA models improve in-domain detection but do not consistently outperform generic models on NFT instances, with F1-score declines of -0.0057 to -0.0151 on non-finance subsets. Practical Implications: Financial institutions and fintech platforms should deploy domain-adapted detection for email-based threats, where finance-specific linguistic cues yield measurable gains, while maintaining generic or ensemble models for broader webpage phishing coverage. Originality/Value: This study introduces a finance-themed, multi-modal, explainable AI framework for phishing detection, demonstrating that domain adaptation depends critically on data modality and feature representation. It provides a novel systematic comparison of generic versus FA phishing detection across both modalities with statistical validation and explainability analysis.

21 May 2026 in Preprints.org
Verified
Preprint
Open access
PDF file
IBIstiaque Bhuiyan
Istiaque Bhuiyan
Muhammad Haseeb AhmadMuhammad Haseeb Ahmad
Muhammad Haseeb Ahmad
AHAriful Hoque
Ariful Hoque
Tanvir BhuiyanTanvir Bhuiyan
Tanvir Bhuiyan
This study examines customer retention intention in neobanking environments using a theory-informed explainable machine learning framework. Existing digital banking research typically relies on linear modelling approaches to explain retention behaviour, potentially overlooking nonlinear, value-range-dependent, and interaction-based predictive patterns. Using a publicly available survey of 305 neobank users, this study compares regularized linear models, a partial least squares structural equation modelling (PLS-SEM)-inspired benchmark, and XGBoost under repeated nested cross-validation. SHapley Additive exPlanations (SHAP)-based explainability, SHAP interaction analysis, generalized additive model (GAM) diagnostics, construct-level aggregation, and construct-sensitivity checks are used to interpret model behaviour and assess robustness. The results show that XGBoost substantially outperforms the linear benchmarks, achieving the lowest average RMSE and highest average R² across 100 out-of-sample test-fold estimates. Trust-related indicators provide the largest share of model-based predictive importance, followed by perceived security and switching costs. SHAP and GAM diagnostics suggest that trust and switching costs may contribute to retention intention in heterogeneous and nonlinear ways, while perceived security displays a more stable positive predictive pattern. Age-related nonlinearities appear weak and should be interpreted cautiously given the young sample profile. The analysis also suggests possible non-additive relationships between trust and perceived security. The study contributes to digital banking and FinTech research by showing how explainable machine learning can complement theory-driven retention models, identify potentially nonlinear predictive patterns, and preserve interpretability. The findings offer practical insight for trust-building, visible security assurance, and retention diagnostics in neobanking contexts.

20 May 2026 in Preprints.org
Verified
Preprint
Open access
PDF file
IBIstiaque Bhuiyan
Istiaque Bhuiyan
Tanvir BhuiyanTanvir Bhuiyan
Tanvir Bhuiyan
Phishing detection models often report strong benchmark performance, yet their reliability under realistic deployment conditions remains uncertain. This study examines this problem by investigating three failure modes of cross-dataset phishing email detection: corpus generalisation failure, asymmetric prevalence-shift failure, and artifact-driven spurious learning. Using six public email corpora, CEAS_08, Enron, Ling, Nazario, Nigerian Fraud, and SpamAssassin, the study evaluates Term Frequency (TF) and Inverse Document Frequency (IDF)-based Logistic Regression and Linear Support Vector Classifier (SVC) models across pooled baseline testing, single-corpus cross-dataset transfer, leave-one-corpus-out pooled training, prevalence-shift simulation, training-prevalence manipulation, dataset-identification analysis, top-feature inspection, artifact-removal ablation, and targeted artifact masking. The findings show that single-corpus models are unstable under cross-dataset transfer, with F1-scores varying substantially across source–target combinations. In contrast, leave-one-corpus-out pooled training improves robustness, with Logistic Regression achieving sustained F1-scores between 0.8201 and 0.8994, and Linear SVC achieving F1-scores between 0.7607 and 0.8910 across unseen corpora. Prevalence-shift experiments reveal that failure is asymmetric and threshold-dependent. High-prevalence-trained models maintain high recall under fixed thresholds but suffer sharp recall degradation when operational alert-budget constraints are imposed. Conversely, low-prevalence-trained models become overly conservative in high-threat environments, producing high precision but substantially lower recall and poorer calibration. Artifact analyses further show that source corpus identity is highly learnable, with dataset-identification accuracy reaching 0.9722 for Logistic Regression and 0.9806 for Linear SVC. Top-feature and masking analyses indicate that models rely partly on corpus markers, date tokens, URL/domain terms, headers, and other artifact-like features rather than only general phishing indicators. The study contributes a deployment-aware and adversary-aware evaluation framework for phishing detection. It shows that benchmark accuracy alone is insufficient for assessing real-world robustness and that reliable phishing detection requires cross-corpus validation, prevalence-aware thresholding, and systematic testing for artifact-driven spurious learning.

2 July 2026 in FinTech
Verified
Journal Article
Open access
PDF file
IBIstiaque Bhuiyan
Istiaque Bhuiyan
Tanvir BhuiyanTanvir Bhuiyan
Tanvir Bhuiyan
Purpose: This study examines whether finance-adapted (FA) phishing detection models improve the detection of finance-themed (FT) attacks, whether improvements differ across email and webpage modalities, and whether finance adaptation creates a specialisation–generalisation trade-off. Design/Methodology/Approach: A domain-aware framework is developed using email (82,486 instances) and webpage (11,430 instances) datasets. FT and non-finance-themed (NFT) instances are identified using weighted lexicon-based labelling. Generic models are compared with FA models across Logistic Regression, Linear SVC, and Random Forest using F1-score, MCC, balanced accuracy, ROC-AUC, and PR-AUC. Statistical validation employs bootstrap confidence intervals and McNemar’s test, while SHAP and permutation importance interpret webpage model behaviour. Findings: FA models outperform generic models in FT email classification, confirming that finance-specific semantic cues improve detection. However, gains are weaker and less consistent in webpage classification, where models rely mainly on structural indicators (page rank, Google index, hyperlinks). The results reveal a specialisation–generalisation trade-off: FA models improve in-domain detection but do not consistently outperform generic models on NFT instances, with F1-score declines of −0.0057 to −0.0151 on non-finance subsets. Practical Implications: Financial institutions and fintech platforms should deploy domain-adapted detection for email-based threats, where finance-specific linguistic cues yield measurable gains, while maintaining generic or ensemble models for broader webpage phishing coverage. Originality/Value: This study introduces a finance-themed, multi-modal, explainable AI framework for phishing detection, demonstrating that domain adaptation depends critically on data modality and feature representation. It provides a novel systematic comparison of generic versus FA phishing detection across both modalities with statistical validation and explainability analysis.

Verified
Journal Article
Open access
PDF file
IBIstiaque Bhuiyan
Istiaque Bhuiyan
Tanvir BhuiyanTanvir Bhuiyan
Tanvir Bhuiyan
Phishing detection models often report strong benchmark performance, yet their reliability under realistic deployment conditions remains uncertain. This study examines this problem by investigating three failure modes of cross-dataset phishing email detection: corpus generalization failure, asymmetric prevalence-shift failure, and artifact-driven spurious learning. Using six public email corpora, CEAS_08, Enron, Ling, Nazario, Nigerian Fraud, and SpamAssassin, the study evaluates Term Frequency (TF) and Inverse Document Frequency (IDF)-based Logistic Regression and Linear Support Vector Classifier (SVC) models across pooled baseline testing, single-corpus cross-dataset transfer, leave-one-corpus-out pooled training, prevalence-shift simulation, training prevalence manipulation, dataset-identification analysis, top-feature inspection, artifact-removal ablation, and targeted feature-sensitivity masking. The findings show that single-corpus models are unstable under cross-dataset transfer, with F1-scores varying substantially across source–target combinations. In contrast, leave-one-corpus-out pooled training improves robustness, with Logistic Regression achieving sustained F1-scores between 0.8201 and 0.8994, and Linear SVC achieving F1-scores between 0.7607 and 0.8910 across unseen corpora. Prevalence-shift experiments reveal that failure is asymmetric and threshold-dependent. High-prevalence-trained models maintain high recall under fixed thresholds but suffer sharp recall degradation when operational alert-budget constraints are imposed. Conversely, low-prevalence-trained models become overly conservative in high-threat environments, producing high precision but substantially lower recall and poorer calibration. Artifact analyses further show that source corpus identity is highly learnable, with dataset-identification accuracy reaching 0.9722 for Logistic Regression and 0.9806 for Linear SVC. Top-feature and masking analyses indicate that models rely partly on corpus markers, date tokens, URL/domain terms, headers, and other artifact-like features rather than only general phishing indicators. The study contributes a deployment-aware and adversary-aware evaluation framework for phishing detection. It shows that benchmark accuracy alone is insufficient for assessing real-world robustness and that reliable phishing detection requires cross-corpus validation, prevalence-aware thresholding, and systematic testing for artifact-driven spurious learning.