<![CDATA[Stock is one of the most popular investment instruments due to its potential to generate substantial returns. However, the high volatility of stock prices requires investors to employ accurate prediction models to support investment decision-making. This study aims to compare the performance of the Artificial Neural Network (ANN) and Support Vector Regression (SVR) methods in predicting the stock price of PT Gudang Garam Tbk using historical data enriched with technical indicators. The study adopted the Cross-Industry Standard Process for Data Mining (CRISP-DM) methodology, which consists of six stages: business understanding, data understanding, data preparation, modeling, evaluation, and deployment. The prediction models were developed using historical stock price data enriched with technical indicators and evaluated using Mean Squared Error (MSE), Root Mean Squared Error (RMSE), Mean Absolute Error (MAE), Mean Absolute Percentage Error (MAPE), and the coefficient of determination (R²). The experimental results demonstrate that the ANN model outperformed the SVR model, achieving an MSE of 2923.86, RMSE of 1709.93, MAE of 1294.76, MAPE of 8.38%, and an R² of 0.68, while the SVR model obtained an MSE of 5211.06, RMSE of 2284.57, MAE of 2126.84, MAPE of 12.73%, and an R² of 0.42. Furthermore, the 240-day forecasting results indicate that the ANN model projects an upward (bullish) trend, whereas the SVR model predicts a relatively stable (sideways) trend. These findings indicate that the Artificial Neural Network (ANN) is more effective than Support Vector Regression (SVR) for predicting the stock price of PT Gudang Garam Tbk, as it produces lower prediction errors and demonstrates superior predictive performance. Keyword: Stock Price Prediction, Artificial Neural Network, Support Vector Regression, CRISP-DM. Abstrak Saham merupakan salah satu instrumen investasi yang banyak diminati karena berpotensi memberikan keuntungan yang tinggi. Namun, tingginya volatilitas harga saham menyebabkan investor memerlukan model prediksi yang akurat sebagai dasar pengambilan keputusan investasi. Penelitian ini bertujuan untuk membandingkan kinerja metode Artificial Neural Network (ANN) dan Support Vector Regression (SVR) dalam memprediksi harga saham PT Gudang Garam Tbk menggunakan data historis yang diperkaya dengan indikator teknikal. Penelitian ini menerapkan metodologi Cross-Industry Standard Process for Data Mining (CRISP-DM) yang meliputi tahapan business understanding, data understanding, data preparation, modeling, evaluation, dan deployment. Model dibangun menggunakan data historis harga saham yang diperkaya dengan indikator teknikal, kemudian dievaluasi menggunakan metrik Mean Squared Error (MSE), Root Mean Squared Error (RMSE), Mean Absolute Error (MAE), Mean Absolute Percentage Error (MAPE), dan koefisien determinasi (R²). Hasil penelitian menunjukkan bahwa model ANN memberikan performa yang lebih baik dibandingkan SVR dengan nilai MSE sebesar 2923,86, RMSE sebesar 1709,93, MAE sebesar 1294,76, MAPE sebesar 8,38%, dan R² sebesar 0,68, sedangkan model SVR memperoleh nilai MSE sebesar 5211,06, RMSE sebesar 2284,57, MAE sebesar 2126,84, MAPE sebesar 12,73%, dan R² sebesar 0,42. Pada prediksi jangka panjang selama 240 hari, model ANN memproyeksikan tren harga yang meningkat (bullish), sedangkan model SVR menghasilkan tren yang relatif stabil (sideways). Berdasarkan hasil tersebut, dapat disimpulkan bahwa metode Artificial Neural Network (ANN) lebih efektif dibandingkan Support Vector Regression (SVR) dalam memprediksi harga saham PT Gudang Garam Tbk karena mampu menghasilkan tingkat kesalahan yang lebih rendah dan kemampuan prediksi yang lebih baik. Kata kunci: Prediksi harga saham; Artificial Neural Network; Support Vector Regression; CRISP-DM.]]>
