Integration of a Hybrid Model with XAI for Stock Price Forecasting: A Dashboard-Driven Approach

This paper on Integration of a Hybrid Model with XAI for Stock price Forecasting focuses on improving the accuracy and transparency of stock price predictions through a combined model of Long Short-Term Memory (LSTM) and MultiProphet, with an emphasis on explainable artificial intelligence (XAI). LSTM which is a deep learning model, is good at identifying patterns in time-series data. while MultiProphet which is an enhanced version of Facebook's Prophet model, effectively handles seasonal variations and anomalies in financial data. By integrating both models, the hybrid approach delivers more reliable stock forecasting. In addition, this project integrates XAI techniques like SHAP and LIME to offer greater insight into how the model generates predictions, which is crucial for informed financial decision-making. Dynamic dashboard provides an intuitive interface for users to track stock price trends, view forecasts, and access explanations of the model's predictions in real-time. This combination of improved prediction accuracy and model transparency enables better decision-making by stakeholders, who can base their actions on clear and interpretable information. Overall, the project represents a significant advancement in making advanced AI models more accessible and actionable in practical, user-oriented financial applications.