An intelligent decision support framework for financial market risk using big data and optimised XGBoost

Risk forecasting in financial markets is essential in management and engineering choices.In this paper, a more refined XGBoost model that uses big data analytics to identify nonlinear patterns that are usually not detected by the conventional technique is developed.The process can be optimised with a high feature selection, hyperparameter optimisation, as well as advanced preprocessing of the data.The model has been tested and proven to have better performance than the current methods in the market and helps in efficient risk management, resource deployment, and timely risk mitigation.The model is responsive to market changes and is ethical on AI operations in real-time.Combined with live data feeds, it will be able to respond more effectively and promptly to changes, a key that can serve as a powerful financial risk prediction instrument, driven by AI.