DETECTING AND PREDICTING FINANCIAL FRAUD USING EXPLAINABLE AI & GRAPH BASED ANOMALY DETECTION

ABSTRACT This literature paper review intends to highlight the shortcomings of traditional fraud detection methods, which often lack adaptability and transparency. Explainable AI (XAI) offers solutions by making model predictions more interpretable, enabling clearer insights for users. Additionally, graph-based anomaly detection captures complex relationships between financial entities, identifying patterns missed by traditional approaches. By integrating XAI with graph-based techniques, this review demonstrates improved accuracy and interpretability in fraud detection, offering advancements in financial auditing and compliance. Keywords— Financial statement fraud, Explainable AI (XAI), Graph-based anomaly detection, Fraud detection, Machine learning.