Year: 2026
Type: article
Abstract: Corporate insolvency poses a significant threat to financial institutions, investors, and overall economic stability. Traditional bankruptcy prediction models, such as logistic regression and Altman's Z-score, often fail to capture complex financial patterns and deliver accurate forecasts. This study proposes the integration of Artificial Intelligence (AI) into Decision Support Systems (DSS) to enhance the ability of businesses and financial institutions to detect insolvency risk in real time. We examine and compare advanced machine learning techniques, including Random Forest, XGBoost, and Long ShortTerm Memory (LSTM) neural networks, to develop a robust insolvency prediction model. Financial data from diverse companies is processed using feature engineering and big data analytics to identify key indicators that contribute to insolvency prediction. The models are evaluated using metrics such as Precision, Recall, F1-score, and overall accuracy, achieving results exceeding 90% in accuracy. The findings demonstrate that AI-driven DSS can support investors, auditors, and financial managers in making datainformed decisions while mitigating financial risk. Additionally, the proposed system functions as an early warning tool by continuously analyzing financial statements and recommending preventive actions for high-risk companies. This study advances the intersection of finance and technology by illustrating how AI can transform insolvency prediction and offer substantial benefits in risk management and market stability. 132 EUROMEDITERRANEAN 23
Source: EUROMEDITERRANEAN
Authors: Sonila Nikolla, Greta Angjeli, Eleina Qirici
Language: English
Subfield: Accounting
Domain: Social Sciences
Open Access status: bronze
OpenAlex ID: W7225315219