Based on big data mining technology, this paper studies the construction and application of enterprise financial risk analysis and prediction model. By comparing multiple machine learning algorithms, this paper constructs an efficient financial risk prediction model and applies it to the risk analysis of enterprise financial data. The research results show that support vector machine (SVM) performs best in evaluation indicators such as accuracy and F1 score, while random forest and neural network can also provide good prediction results in certain specific cases. Through visualization methods such as box plots, pairwise relationship diagrams and violin plots, the intrinsic relationship between financial indicators and enterprise financial risks is further revealed. This study provides scientific decision-making support for enterprise financial management and demonstrates the wide application potential of big data technology in the field of risk prediction.