IERSE: An Explainable Hybrid Eligibility and Risk Scoring Framework for Immigration Decision Support

Immigration adjudication is a complex, high-stakes process involving detailed eligibility criteria, extensive documentation, and multi-factor risk assessment. Human officers retain full decision authority, but manual preliminary evaluation is time-consuming and prone to inconsistencies. This paper introduces the Intelligent Eligibility and Risk Scoring Engine (IERSE), a humanin-the-loop, explainable decision-support framework combining rule-based eligibility evaluation with machine learning-based multi-dimensional risk scoring. IERSE processes structured and unstructured applicant data, applies deterministic eligibility rules, and generates interpretable risk indicators across fraud, compliance, identity, economic, and security dimensions. Simulation results on 10,000 synthetic profiles demonstrate the potential to reduce preliminary processing time by approximately 65% relative to fully manual preliminary screening, while achieving high consistency in eligibility evaluation (98%) and strong precision and recall for risk signal identification. The framework is explicitly designed to support officer judgment rather than automate adjudication, providing transparent, auditable outputs that assist case prioritization and review. IERSE illustrates how hybrid, explainable AI architectures can improve efficiency, consistency, and transparency in high-stakes public-sector decision-support workflows while preserving human authority.