Auditable Climate Risk Intelligence from Fragmented ESG Data: Deterministic Orchestration and Imbalance-Aware Learning for Scope 1-3 Validation

This paper proposes an auditable climate risk intelligence framework for validating fragmented ESG data, integrating deterministic orchestration, temporal anomaly detection, and imbalance-aware ensemble learning. It introduces a synthetic ESG validation benchmark to support open reproducibility.