INTELLIGENT METHODS FOR ASSESSING ENTERPRISE SOLVENCY: MACHINE LEARNING VS TRADITIONAL FINANCIAL ANALYSIS

In an increasingly complex financial environment, assessing enterprise solvency has become a critical task for risk management and decision-making. This article presents a comparative analysis of traditional financial analysis methods and intelligent machine learning approaches used to evaluate corporate solvency. The study covers classical tools such as ratio analysis and integrated scoring models, alongside machine learning algorithms including logistic regression, random forest, XGBoost, and neural networks. The comparison focuses on predictive accuracy, interpretability, data requirements, and applicability across use cases. The article demonstrates that while machine learning offers enhanced adaptability and precision, its effectiveness depends on data quality and explainability mechanisms. Recommendations are formulated on how to select an appropriate method depending on the purpose of the analysis, data availability, resource constraints, and level of system integration. The study concludes with the argument for hybrid frameworks that combine the strengths of both traditional and intelligent models.