Building an Intelligent Model for Identifying Corporate Financial Fraud by Integrating Big Data Mining and Machine Learning Technologies

Intelligent anomaly detection based on big data analytics and machine learning has become an essential approach for data-driven monitoring and decision support in complex engineering systems. To improve the efficiency and reliability of corporate financial fraud identification, this study develops an intelligent detection framework that integrates multisource data mining, feature engineering, ensemble learning, and dynamic model optimization. The proposed methodology combines heterogeneous financial information with machine learning algorithms to construct adaptive fraud identification models capable of capturing hidden abnormal patterns and evolving fraud behaviors. A systematic strategy involving data integration, feature selection, algorithm fusion, incremental model updating, and interpretability enhancement is further presented to improve detection robustness and practical applicability. The framework effectively addresses challenges associated with data quality, sample imbalance, feature adaptability, and model transparency while supporting continuous optimization through human–machine collaborative feedback. Beyond financial risk management, the proposed intelligent identification architecture provides methodological references for anomaly detection, distributed data fusion, adaptive pattern recognition, and intelligent monitoring in engineering systems, offering potential applications for information processing and decision support in Electromagnetic Waves, Antennas and Propagation, particularly in wireless sensing networks and data-driven signal analysis.