Advanced Computing–Enabled Secure Financial Information Systems for Real-Time Fraud Detection in U.S. Digital Payments: A Quantitative Analysis

Abstract

This study addressed the persistent problem that U.S. digital payment ecosystems require fraud detection decisions in real time, yet many organizations struggle to align scalable advanced computing with secure financial information system controls in ways that measurably improve fraud detection effectiveness while remaining auditable and operationally usable. The purpose was to quantify how advanced computing enablement (ACE) and secure control strength (SCS) predict real-time fraud detection effectiveness (RTFDE), and to examine whether alert trust and actionability (ATA) strengthens practical effectiveness within enterprise and cloud-based operational cases. Using a quantitative, cross-sectional, case-based design, data were collected from a multi-role sample of enterprise digital-payment stakeholders (N = 210), including fraud and risk operations (38.1%), IT and data engineering (34.3%), and security and compliance (27.6%), with mean experience of 6.8 years (SD = 3.9). Key variables were ACE (8 items), SCS (9 items), RTFDE (8 items), ATA (6 items), feature adoption and maturity (FAM), and operational error hotspots (OEH), all measured on 5-point Likert scales with strong internal consistency (ACE α = .88; SCS α = .91; RTFDE α = .89; ATA α = .86). The analysis plan applied descriptive statistics, Pearson correlations, and hierarchical multiple regression with diagnostic checks (VIF 1.31–2.08). Headline findings showed moderately high capability ratings (ACE M = 3.74, SD = 0.61; SCS M = 3.88, SD = 0.55; RTFDE M = 3.69, SD = 0.63) and strong positive associations with effectiveness (ACE–RTFDE r = .62, p < .001; SCS–RTFDE r = .58, p < .001). Regression results indicated ACE alone explained 32% of variance in RTFDE (R² = .32; β = .57, p < .001); adding SCS increased explanatory power to 40% (ΔR² = .08; ACE β = .41, p < .001; SCS β = .29, p < .001); adding ATA raised total explained variance to 44% (ΔR² = .04; ACE β = .32, p < .001; SCS β = .21, p = .002; ATA β = .24, p < .001). Operationally, the largest constraints clustered in data quality and feature completeness (HS = 3.92), alert routing and prioritization (HS = 3.71), and manual review capacity (HS = 3.63), implying that improving feature integrity and workflow throughput can yield gains even where baseline security controls are mature. Implications suggest that payment organizations should jointly invest in real-time analytics maturity and security traceability, while optimizing alert usability to reduce false-positive burden and strengthen defensible decision trails.