Investment decision-making in the capital market requires accurate analysis to minimize risk and maximize returns. This study focuses on PT Unilever Indonesia Tbk, a leading Fast-Moving Consumer Goods (FMCG) company listed on the Indonesia Stock Exchange, to evaluate its stock price prediction using the linear regression method supported by RapidMiner software. Historical stock data were collected from January 2, 2018, to June 27, 2023, including attributes such as opening price, closing price, lowest price, highest price, and trading volume. The dataset was processed using data screening and modeling techniques to construct a linear regression model for prediction. Various scenarios with different proportions of training and testing data (70/30, 80/20, 60/40, and 90/10) were tested to analyze the impact of data distribution on model performance. The evaluation results showed that the 80% training and 20% testing scenario provided the lowest Root Mean Squared Error (RMSE) of 56.699, indicating better predictive accuracy compared to other scenarios. Nevertheless, the linear regression model still produced a relatively high error rate, with the best RMSE value suggesting limitations of this approach for complex market prediction. This study concludes that while linear regression can provide a basic framework for stock price forecasting, incorporating additional economic, fundamental, and external factors could significantly improve predictive reliability. The findings offer practical insights for investors and researchers in understanding the potential and limitations of linear regression in stock market analysis