A Hybrid CNN Bi-LSTM Framework for Multi-Class Plant Disease Detection and Health Value Estimation

Janani  J.; Gautham R.; Suguna  C.; Ponni  K.

doi:https://doi.org/10.54216/IJAACI.080103

Financial Technology and Innovation | 2024

From Account Access to Payment Value: A Business Readiness Model for FinTech Innovation

Digital finance markets often expand through account ownership before those accounts become active sources of payment value, merchant participation and durable financial behaviour. This paper develops a business-oriented FinTech readiness model that separates access, activation, merchant conversion, stored-value behaviour and resilience. The analysis uses regional and income-group indicators from the Global Findex database to examine how account access is transformed into commercially meaningful digital payment use. The results show that account ownership alone is an incomplete measure of FinTech market opportunity. High-income economies have the strongest overall readiness, East Asia and Pacific shows strong merchant-payment conversion, Sub-Saharan Africa has a distinctive mobile-money channel, and low-income economies show large unmet activation potential. The paper contributes a practical scorecard for banks, payment firms and regulators by showing where digital finance strategy should focus: onboarding, usage activation, merchant acceptance, account-based value retention, or trust and resilience safeguards.

groups

Samandarboy Sulaymanov mail -

Durdona Davletova mail

link https://doi.org/10.54216/FinTech-I.040104

Volume & Issue

Vol. Volume 4 / Iss. Issue 1

Details open_in_new

Financial Technology and Innovation | 2024

The FinTech Maturity Divide: Benchmarking AI Services, Open Banking, and Embedded Finance Across Financial Service Sectors

The global financial services industry is undergoing a structural transformation driven by the convergence of artificial intelligence, open application programming interfaces, digital payment infrastructure, and embedded financial services. Despite sustained investment activity, the empirical evidence base for comparing technology adoption maturity across institutional types and capability domains remains fragmented, leaving executives and policymakers without the benchmarking evidence needed for informed strategic investment decisions. This paper addresses that gap through a systematic multi-dimensional maturity study drawing on primary survey data from financial industry professionals across multiple countries and a consumer adoption survey of retail banking customers, covering five technology dimensions: artificial intelligence services, open banking and application programming interface ecosystems, digital payment infrastructure, embedded finance, and regulatory technology. Challenger banks and FinTech startups substantially outperform traditional incumbent institutions on open banking and embedded finance, while traditional institutions retain a relative advantage in regulatory technology compliance. Payment processors dominate on digital payment maturity but show the widest capability gap in artificial intelligence and embedded finance. Consumer adoption analysis reveals pronounced age-related disparities in buy-now-pay-later, cryptocurrency, and robo-advisory services with direct implications for financial inclusion strategy. Regression analysis identifies application programming interface readiness as the single strongest predictor of overall maturity, confirming that foundational data architecture investment has compounding returns across all five technology domains. The paper contributes a validated five-dimension maturity framework, a regression model of the institutional and strategic predictors of overall FinTech maturity, and ten evidence-based strategic recommendations for executives navigating the technology transformation of financial services.

groups

Davletov I. Rakhimberganovich mail -

Dusmuratov R. Davlatbayevich mail

link https://doi.org/10.54216/FinTech-I.040105

Volume & Issue

Vol. Volume 4 / Iss. Issue 1

Details open_in_new

Financial Technology and Innovation | 2025

Explainable Artificial Intelligence for Real-Time Financial Fraud Detection: A Systematic Literature Review

Financial fraud detection systems increasingly rely on machine learning to identify suspicious transactions at scale. However, the opacity of many high-performing models raises significant concerns regarding trust, regulatory compliance, and practical deployment in real-time financial environments. Explainable Artificial Intelligence (XAI) has emerged as a promising solution to enhance transparency and accountability, yet its feasibility under real-time constraints remains unclear. This systematic literature review examines empirical studies on explainable AI approaches for financial fraud detection, with explicit focus on real-time applicability. Following PRISMA guidelines, nineteen peer-reviewed empirical studies were selected and analyzed based on fraud domain, model type, explainability technique, evaluation metrics, and evidence of real-time performance. Results show that posthoc explanation methods, particularly SHAP and LIME, dominate the literature, while intrinsic explainability and deployment-level latency reporting remain limited. Despite frequent claims of real-time applicability, only one study provides quantitative runtime evidence. The findings highlight critical gaps: absence of explanation latency evaluation, lack of deployment-oriented validation, and insufficient regulatory compliance integration. This review reveals a systematic disconnect between real-time claims and empirical evidence, establishing the need for standardized latency benchmarking in explainable fraud detection research.

groups

Ulugbek Inoyatov mail -

Eugene Q. Castro mail

link https://doi.org/10.54216/FinTech-I.050101

Volume & Issue

Vol. Volume 5 / Iss. Issue 1

Details open_in_new

Financial Technology and Innovation | 2025

From Principles to Practice: A Cross-Sector Assessment of Responsible AI Governance Readiness

The rapid institutionalisation of artificial intelligence across financial services, healthcare, technology, and the public sector has generated a parallel proliferation of governance frameworks, ethical principles, and regulatory instruments that collectively demand organisations translate abstract values into operational practice. The gap between stated principle and enacted governance—what we term the responsible AI implementation gap—is now recognised as one of the central practical challenges in AI deployment, yet its magnitude, distribution across sectors, and organizational determinants remain poorly characterised in the empirical literature. This paper addresses that gap through a three-phase mixed-methods programme combining systematic analysis of publicly available governance frameworks, a cross-sector practitioner survey, and a governance maturity scoring exercise. Significant variation is documented across sectors on all five governance dimensions examined, with the technology sector leading on accountability and transparency, healthcare on privacy and human oversight, and the public sector on regulatory compliance readiness. Across all sectors, however, a persistent and pronounced gap exists between the governance principles that organisations formally endorse and the operational processes through which those principles are enacted: the average policy-to-practice gap across all eight governance principles assessed is consistent and substantial. Regression analysis identifies the presence of a dedicated responsible AI team as the single strongest organisational predictor of maturity, followed by staff training investment and senior executive sponsorship. The paper contributes a validated governance maturity framework, a framework coverage taxonomy for twenty-four public AI governance instruments, and six evidence-based implementation guidelines for organisations seeking to move from principle adoption to genuine operational accountability.

groups

Mahmoud A. Zaher mail

link https://doi.org/10.54216/FinTech-I.050102

Volume & Issue

Vol. Volume 5 / Iss. Issue 1

Details open_in_new

Financial Technology and Innovation | 2025

Credit-Card Engagement Segmentation for Embedded FinTech Product Strategy: Evidence from Expenditure Microdata

Digital card products create business value only when issued accounts translate into sustained, responsible use. This paper develops an expenditure-based segmentation model for embedded FinTech card strategy using a real credit-card micro-dataset. Average monthly card expenditure is treated as an observable engagement outcome and is examined alongside income, age, and home-ownership status. The empirical design combines descriptive portfolio profiling, robust regression, cross-validated prediction, and product-action mapping. The results show that income is the strongest observed driver of monthly spend, but the relationship is nonlinear and does not fully explain customer heterogeneity. A quartile-based segmentation separates low-use, developing-use, active-use, and premium-use customers, with mean monthly expenditure increasing from 39.92 to 666.35 across the four operating segments. The analysis argues that card engagement should be managed as a portfolio state rather than as a simple activation metric. The study contributes a transparent business-analytics framework that links observed card expenditure to embedded-finance decisions, including activation support, limit calibration, reward design, and repayment-aware engagement monitoring.

groups

Samandarboy Sulaymanov mail -

Olimjonov Abbosjon Olimjonovich mail

link https://doi.org/10.54216/FinTech-I.050103

Volume & Issue

Vol. Volume 5 / Iss. Issue 1

Details open_in_new

Financial Technology and Innovation | 2025

Trustworthy Digital Onboarding Readiness in FinTech Markets: A Business Analytics Model for e-KYC Conversion

Digital onboarding has become a decisive business capability for financial technology firms because customer acquisition, compliance screening, product activation, and trust formation now occur in the same online journey. This paper proposes a Digital Onboarding Readiness model for evaluating whether a market has the conditions required to convert identity verification into sustained FinTech usage. The model combines account access, digital payment use, mobile internet readiness, digital identity support, open-finance policy, regulatory onboarding readiness, and consumer trust into a business-oriented index. A cross-market indicator panel is analysed using descriptive profiling, maturity clustering, readiness decomposition, and predictive interpretation. The results show that strong account ownership alone does not guarantee onboarding maturity. Markets with advanced identity and policy infrastructure may still face low payment-use conversion, while markets with widespread digital payments may be constrained by trust and regulatory readiness gaps. The findings suggest that FinTech firms should treat onboarding as a portfolio capability rather than a front-end compliance step. The paper contributes a transparent measurement framework for market entry, platform partnerships, and responsible e-KYC investment decisions.

groups

Ilknur Ozturk mail

link https://doi.org/10.54216/FinTech-I.050104

Volume & Issue

Vol. Volume 5 / Iss. Issue 1

Details open_in_new

Financial Technology and Innovation | 2025

Beyond the Branch: Consumer Adoption, Satisfaction, and Financial Advisor Acceptance of FinTech Services Across Retail Banking, Challenger, and Wealth Management Segments

The structural transformation of retail financial services by mobile banking platforms, FinTech applications, open banking ecosystems, and AI-powered credit and advisory tools has created both unprecedented opportunities for financial inclusion and a pronounced gap between adoption rates achievable in high-digital-literacy segments and those attainable in mainstream and mass-market contexts. Comparative evidence assessing customer satisfaction and financial advisor acceptance simultaneously across multiple FinTech service domains and institutional segments remains sparse, limiting the evidence base available to practitioners and policymakers designing inclusive FinTech deployment strategies. The present investigation enrolled retail banking customers across four institutional segments—traditional banks, challenger banks, credit unions, and private banking divisions alongside a parallel cohort of relationship managers and financial advisors, to assess adoption rates and satisfaction across four FinTech domains: mobile and digital banking, FinTech financial applications, open banking and personal financial management, and AI-powered credit assessment and advisory services. Significant between-segment variation was documented across all four domains, with private banking customers reporting the highest satisfaction and adoption and credit union customers the lowest. AI-powered credit and advisory services elicited the lowest customer satisfaction across all segments and the largest customer-advisor divergence. Digital financial literacy and prior FinTech experience emerged as the two strongest independent predictors of adoption. The investigation contributes a validated crosssegment measurement instrument, customer-profile-specific adoption profiles, and evidence-based recommendations for financial service providers deploying FinTech capabilities across heterogeneous customer bases.

groups

Serkan Yilmaz Kandir mail -

Murat Ismet Haseki mail

link https://doi.org/10.54216/FinTech-I.050105

Volume & Issue

Vol. Volume 5 / Iss. Issue 1

Details open_in_new

International Journal of Advances in Applied Computational Intelligence | 2026

Motion Vector–Guided Object Detection and Tracking for Smart Surveillance Systems

Multiple moving object detection and tracking are challenging roles in many computer vision applications such as object navigation and human identification. Object tracking is one of the key challenges for securing against crime, supporting public safety, and enabling effective traffic management systems. In video surveillance applications, detection of multiple moving vehicles from video is the major task for tracking and understanding the behavior of the detected objects. Performance of object detection algorithms is degraded by factors such as fog or haze, occlusion, dynamic background, poor illumination, and low resolution. Fog is one of the major bottlenecks of video surveillance applications. The proposed Dark Channel Prior algorithm using guided filter (GDCP) is adapted for fog removal. The Gaussian Mixture Model (GMM) is proposed for detecting multiple moving objects, and features are extracted from the detected objects using Motion Vector Estimation. The K-Nearest Neighbor algorithm is used for tracking the moving objects (vehicles) using the detected features. Efficiency is improved due to the adoption of the proposed fog-removal algorithm and feature extraction for effective tracking. There are wide varieties of applications in moving object detection and tracking.

groups

V. Vinothini mail -

N. Devi mail -

R. Roja mail -

G. Mahendran mail

link https://doi.org/10.54216/IJAACI.080101

Volume & Issue

Vol. Volume 8 / Iss. Issue 1

Details open_in_new

International Journal of Advances in Applied Computational Intelligence | 2026

XAI-DermNet: A Dual-Modality Deep Learning and Explainable AI Fusion Framework for Transparent Skin Lesion Diagnosis from Dermoscopic Images

The advancement of trustworthy diagnostic tools in dermatological automation is hindered by the limited transparency of current deep learning systems, which function as opaque models and impede clinical acceptance. This research presents a novel intelligent framework for skin lesion analysis that unites deep learning methodologies with explainable artificial intelligence (XAI) principles to address this interpretability deficit. The proposed approach utilizes a transfer-learned ResNet50 architecture for robust image classification, coupled with Local Interpretable Model-agnostic Explanations (LIME) to furnish clear, visual justifications for the model’s outputs. Performance assessment on the HAM10000 benchmark yielded a classification accuracy of 94.3%, with a validation accuracy of 91.8%. Concurrently, the LIME framework effectively identified and visualized diagnostically critical features in the lesion images, thereby elucidating the model’s reasoning process for medical practitioners. These findings confirm that augmenting high-performance deep learning with post-hoc explanatory techniques yields a credible and understandable diagnostic instrument, thereby promoting clinician trust and facilitating data-informed medical judgments. Subsequent developments will prioritize scalable cloud implementation, interoperability with healthcare information systems, extension to underrepresented lesion categories, and rigorous evaluation in diverse clinical environments.

groups

Sukkirtha K. mail -

Anbuchelian S. mail -

John A. mail

link https://doi.org/10.54216/IJAACI.080102

Volume & Issue

Vol. Volume 8 / Iss. Issue 1

Details open_in_new

International Journal of Advances in Applied Computational Intelligence | 2026

A Hybrid CNN Bi-LSTM Framework for Multi-Class Plant Disease Detection and Health Value Estimation

Accurate and early identification of plant diseases is essential for ensuring sustainable agriculture and maximizing crop productivity. This paper presents a hybrid deep learning framework integrating Convolutional Neural Networks (CNN) and Bidirectional Long Short-Term Memory (BiLSTM) networks for multi-class plant disease detection, classification, and Plant Health Value (PHV) estimation. The proposed framework begins with a comprehensive data preprocessing pipeline involving image resizing, normalization, and augmentation to improve model robustness. The CNN module extracts critical spatial and visual features such as lesion shape, leaf texture, and color intensity, while the BiLSTM model captures temporal and sequential feature correlations to accurately learn disease progression patterns. A Decision Support System (DSS) is incorporated to compute the Plant Health Value (PHV), where PHV ranges from 0% (Healthy) to 100% (Severely Unhealthy), indicating the severity of disease infection. Additionally, the DSS generates actionable recommendations to assist in early intervention and treatment planning. Experimental results on a multi-species plant dataset demonstrate that the proposed CNN–BiLSTM hybrid model significantly improves accuracy, interpretability, and early disease prediction compared to conventional CNN based methods, offering a robust and intelligent framework for automated plant health monitoring.

groups

Janani J. mail -

Gautham R. mail -

Suguna C. mail -

Ponni K. mail

link https://doi.org/10.54216/IJAACI.080103

Volume & Issue

Vol. Volume 8 / Iss. Issue 1

Details open_in_new