This research presents a hybrid model combining network analysis and machine learning to improve financial distress prediction. Utilizing ensemble learning methods such as voting classifiers and Random Forest algorithms, the model addresses the complexities of predicting financial distress with a dataset encompassing 86 features and 3,672 samples. By integrating cutting-edge machine learning techniques with effective ensemble strategies, the model aims to enhance both accuracy and reliability in identifying financially distressed entities. The study assesses the efficacy of these methods in differentiating between distressed and non-distressed entities, revealing notable improvements in predictive performance. This advancement provides valuable insights for financial analysts and decision-makers, offering a refined approach to understanding financial health indicators.

Keywords: Ensemble learning, voting classifiers and Random Forest algorithms.