Intelligent Accounting System Based on Machine Learning

The accounting sector is currently witnessing a transformative shift, moving away from traditional, labor-intensive processes towards more efficient, automated systems. This study explores the application of machine learning, specifically the XGBoost algorithm, to revolutionize accounting practices. We employed a robust methodology involving data preprocessing, feature engineering, and iterative training of the XGBoost model, focusing on enhancing prediction accuracy and processing efficiency. Our experiments reveal that the XGBoost-based system excels in classification, anomaly detection, and intelligent report generation, significantly outperforming traditional methods. Notably, the model achieved a classification accuracy of 95.8%, an anomaly detection F1-score of 98.85%, and produced intelligent reports within 10 minutes with 99.5% data consistency. These findings underscore the potential of machine learning to not only improve the accuracy and efficiency of financial management but also to redefine the landscape of accounting practices. Future research will aim to refine these models further, ensuring scalability and user accessibility for broader adoption in the industry.