Balasubramanian S and Natarajan M, Annamalai University, India

Crude oil price prediction remains a challenging task due to volatile market conditions. Traditional models often fail to adjust, while modern AI models like LSTM can find patterns but are hard to explain. This study combines Machine Learning (Random Forest), Deep Learning (LSTM), and explainability tools (SHAP and LIME). This study aims to develop a model that is both accurate and easy to understand. The results exhibit that the combined model offers better accuracy and trust for decision-making.

Explainable AI, Trade Forecasting, Random Forest, LSTM, SHAP, LIME, ARIMA, Prophet, Deep Learning, Crude Oil Price Prediction