Predictive Analysis of Gold Prices Using Machine Learning Approaches
Keywords:
Gold Price Prediction, Machine Learning, Deep Learning, LSTM, Natural Language Processing (NLP), Sentiment Analysis, Time Series Forecasting, Explainable AI (XAI), Economic IndicatorsAbstract
Gold is one of the most important financial assets, and accurately predicting its price is a challenging task due to
the influence of various economic and market-related factors. This study proposes an intelligent gold price
prediction system that combines machine learning and deep learning techniques to improve forecasting
performance. The proposed framework integrates structured financial data, including historical gold prices,
inflation rates, interest rates, and exchange rates, with unstructured information such as financial news and social
media sentiment to better capture market trends and investor behavior.
Among the models evaluated, the Long Short-Term Memory (LSTM) model achieved the best performance, with
an accuracy of 94.2%, a Root Mean Square Error (RMSE) of 2.31, and a Mean Absolute Error (MAE) of 1.87,
outperforming traditional machine learning models such as Linear Regression and Support Vector Machine
(SVM). The proposed approach also incorporates sentiment analysis and Explainable Artificial Intelligence
(XAI), enabling not only more accurate predictions but also greater transparency in understanding the factors
influencing the model's decisions. The results demonstrate that the proposed framework is an effective, reliable,
and scalable solution for real-time gold price forecasting and can support investors, analysts, and financial
institutions in making informed decisions.
